Expose a route to show advanced index stats

Merge #4667
4667: Frequency matching strategy r=Kerollmops a=ManyTheFish # Pull Request ## Related issue Fixes #3773 ## What does this PR do? - add test for matching strategy - implement frequency matching strategy See the [PRD for more details](https://www.notion.so/meilisearch/Frequency-Matching-Strategy-0f3ba08833a442a39590a53a1505ab00). [Public API](https://www.notion.so/meilisearch/frequency-matching-strategy-89868fb7fc584026bc56e378eb854a7f). Co-authored-by: ManyTheFish <many@meilisearch.com>
2025-07-18 04:11:07 +00:00 · 2024-06-04 14:08:01 -04:00 · 2024-05-30 14:53:31 +00:00 · 2024-05-30 12:02:42 +02:00 · 2024-05-29 16:55:20 +00:00 · 2024-05-29 17:48:48 +02:00
124 changed files with 8358 additions and 1213 deletions
--- a/.github/workflows/bench-pr.yml
+++ b/.github/workflows/bench-pr.yml
@ -43,4 +43,11 @@ jobs:

        - name: Run benchmarks on PR ${{ github.event.issue.id }}
          run: |
-            cargo xtask bench --api-key "${{ secrets.BENCHMARK_API_KEY }}" --dashboard-url "${{ vars.BENCHMARK_DASHBOARD_URL }}" --reason "[Comment](${{ github.event.comment.html_url }}) on [#${{ github.event.issue.number }}](${{ github.event.issue.html_url }})" -- ${{ steps.command.outputs.command-arguments }}
+            cargo xtask bench --api-key "${{ secrets.BENCHMARK_API_KEY }}" \
+               --dashboard-url "${{ vars.BENCHMARK_DASHBOARD_URL }}" \
+               --reason "[Comment](${{ github.event.comment.html_url }}) on [#${{ github.event.issue.number }}](${{ github.event.issue.html_url }})" \
+               -- ${{ steps.command.outputs.command-arguments }} > benchlinks.txt
+
+        - name: Send comment in PR
+          run: |
+            gh pr comment ${{github.event.issue.number}} --body-file benchlinks.txt
--- a/BENCHMARKS.md
+++ b/BENCHMARKS.md
@ -187,8 +187,8 @@ They are JSON files with the following structure (comments are not actually supp
  },
  // Core of the workload.
  // A list of commands to run sequentially.
-  // A command is a request to the Meilisearch instance that is executed while the profiling runs.
-  "commands": [
+  // Optional: A precommand is a request to the Meilisearch instance that is executed before the profiling runs.
+  "precommands": [
    {
      // Meilisearch route to call. `http://localhost:7700/` will be prepended.
      "route": "indexes/movies/settings",
@ -224,8 +224,11 @@ They are JSON files with the following structure (comments are not actually supp
      // - DontWait: run the next command without waiting the response to this one.
      // - WaitForResponse: run the next command as soon as the response from the server is received.
      // - WaitForTask: run the next command once **all** the Meilisearch tasks created up to now have finished processing.
-      "synchronous": "DontWait"
-    },
+      "synchronous": "WaitForTask"
+    }
+  ],
+  // A command is a request to the Meilisearch instance that is executed while the profiling runs.
+  "commands": [
    {
      "route": "indexes/movies/documents",
      "method": "POST",
--- a/Cargo.lock
+++ b/Cargo.lock
@ -378,9 +378,9 @@ dependencies = [

 [[package]]
 name = "arroy"
-version = "0.2.0"
+version = "0.3.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "efddeb1e7c32a551cc07ef4c3e181e3cd5478fdaf4f0bd799983171c1f6efe57"
+checksum = "73897699bf04bac935c0b120990d2a511e91e563e0f9769f9c8bb983d98dfbc9"
 dependencies = [
 "bytemuck",
 "byteorder",
@ -500,7 +500,7 @@ checksum = "8c3c1a368f70d6cf7302d78f8f7093da241fb8e8807c05cc9e51a125895a6d5b"

 [[package]]
 name = "benchmarks"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "bytes",
@ -645,7 +645,7 @@ dependencies = [

 [[package]]
 name = "build-info"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "time",
@ -1536,16 +1536,16 @@ checksum = "fea41bba32d969b513997752735605054bc0dfa92b4c56bf1189f2e174be7a10"

 [[package]]
 name = "doxygen-rs"
-version = "0.2.2"
+version = "0.4.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bff670ea0c9bbb8414e7efa6e23ebde2b8f520a7eef78273a3918cf1903e7505"
+checksum = "415b6ec780d34dcf624666747194393603d0373b7141eef01d12ee58881507d9"
 dependencies = [
 "phf",
 ]

 [[package]]
 name = "dump"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "big_s",
@ -1793,7 +1793,7 @@ dependencies = [

 [[package]]
 name = "file-store"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "faux",
 "tempfile",
@ -1816,7 +1816,7 @@ dependencies = [

 [[package]]
 name = "filter-parser"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "insta",
 "nom",
@ -1836,7 +1836,7 @@ dependencies = [

 [[package]]
 name = "flatten-serde-json"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "criterion",
 "serde_json",
@ -1954,7 +1954,7 @@ dependencies = [

 [[package]]
 name = "fuzzers"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "arbitrary",
 "clap",
@ -2262,12 +2262,11 @@ checksum = "95505c38b4572b2d910cecb0281560f54b440a19336cbbcb27bf6ce6adc6f5a8"

 [[package]]
 name = "heed"
-version = "0.20.0-alpha.9"
+version = "0.20.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9648a50991c86df7d00c56c268c27754fcf4c80be2ba57fc4a00dc928c6fe934"
+checksum = "6f7acb9683d7c7068aa46d47557bfa4e35a277964b350d9504a87b03610163fd"
 dependencies = [
 "bitflags 2.5.0",
- "bytemuck",
 "byteorder",
 "heed-traits",
 "heed-types",
@ -2281,15 +2280,15 @@ dependencies = [

 [[package]]
 name = "heed-traits"
-version = "0.20.0-alpha.9"
+version = "0.20.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5ab0b7d9cde969ad36dde692e487dc89d97f7168bf6a7bd3b894ad4bf7278298"
+checksum = "eb3130048d404c57ce5a1ac61a903696e8fcde7e8c2991e9fcfc1f27c3ef74ff"

 [[package]]
 name = "heed-types"
-version = "0.20.0-alpha.9"
+version = "0.20.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f0cb3567a7363f28b597bf6e9897b9466397951dd0e52df2c8196dd8a71af44a"
+checksum = "3cb0d6ba3700c9a57e83c013693e3eddb68a6d9b6781cacafc62a0d992e8ddb3"
 dependencies = [
 "bincode",
 "byteorder",
@ -2448,7 +2447,7 @@ checksum = "206ca75c9c03ba3d4ace2460e57b189f39f43de612c2f85836e65c929701bb2d"

 [[package]]
 name = "index-scheduler"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "big_s",
@ -2465,7 +2464,6 @@ dependencies = [
 "meilisearch-auth",
 "meilisearch-types",
 "page_size 0.5.0",
- "puffin",
 "rayon",
 "roaring",
 "serde",
@ -2644,7 +2642,7 @@ dependencies = [

 [[package]]
 name = "json-depth-checker"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "criterion",
 "serde_json",
@ -3189,14 +3187,13 @@ checksum = "f9d642685b028806386b2b6e75685faadd3eb65a85fff7df711ce18446a422da"

 [[package]]
 name = "lmdb-master-sys"
-version = "0.1.0"
+version = "0.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "629c123f5321b48fa4f8f4d3b868165b748d9ba79c7103fb58e3a94f736bcedd"
+checksum = "dc9048db3a58c0732d7236abc4909058f9d2708cfb6d7d047eb895fddec6419a"
 dependencies = [
 "cc",
 "doxygen-rs",
 "libc",
- "pkg-config",
 ]

 [[package]]
@ -3233,12 +3230,6 @@ version = "0.4.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "90ed8c1e510134f979dbc4f070f87d4313098b704861a105fe34231c70a3901c"

-[[package]]
-name = "lz4_flex"
-version = "0.10.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8b8c72594ac26bfd34f2d99dfced2edfaddfe8a476e3ff2ca0eb293d925c4f83"
-
 [[package]]
 name = "macro_rules_attribute"
 version = "0.2.0"
@ -3281,7 +3272,7 @@ checksum = "490cc448043f947bae3cbee9c203358d62dbee0db12107a74be5c30ccfd09771"

 [[package]]
 name = "meili-snap"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "insta",
 "md5",
@ -3290,7 +3281,7 @@ dependencies = [

 [[package]]
 name = "meilisearch"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "actix-cors",
 "actix-http",
@ -3343,7 +3334,6 @@ dependencies = [
 "pin-project-lite",
 "platform-dirs",
 "prometheus",
- "puffin",
 "rand",
 "rayon",
 "regex",
@ -3383,7 +3373,7 @@ dependencies = [

 [[package]]
 name = "meilisearch-auth"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "base64 0.21.7",
 "enum-iterator",
@ -3402,7 +3392,7 @@ dependencies = [

 [[package]]
 name = "meilisearch-types"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "actix-web",
 "anyhow",
@ -3432,7 +3422,7 @@ dependencies = [

 [[package]]
 name = "meilitool"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "clap",
@ -3471,7 +3461,7 @@ dependencies = [

 [[package]]
 name = "milli"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "arroy",
 "big_s",
@ -3511,7 +3501,6 @@ dependencies = [
 "obkv",
 "once_cell",
 "ordered-float",
- "puffin",
 "rand",
 "rand_pcg",
 "rayon",
@ -3912,7 +3901,7 @@ checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"

 [[package]]
 name = "permissive-json-pointer"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "big_s",
 "serde_json",
@ -4182,23 +4171,6 @@ version = "2.28.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "106dd99e98437432fed6519dedecfade6a06a73bb7b2a1e019fdd2bee5778d94"

-[[package]]
-name = "puffin"
-version = "0.16.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "76425abd4e1a0ad4bd6995dd974b52f414fca9974171df8e3708b3e660d05a21"
-dependencies = [
- "anyhow",
- "bincode",
- "byteorder",
- "cfg-if",
- "instant",
- "lz4_flex",
- "once_cell",
- "parking_lot",
- "serde",
-]
-
 [[package]]
 name = "pulp"
 version = "0.18.9"
@ -6080,7 +6052,7 @@ dependencies = [

 [[package]]
 name = "xtask"
-version = "1.8.0"
+version = "1.9.0"
 dependencies = [
 "anyhow",
 "build-info",
--- a/Cargo.toml
+++ b/Cargo.toml
@ -22,7 +22,7 @@ members = [
 ]

 [workspace.package]
-version = "1.8.0"
+version = "1.9.0"
 authors = [
    "Quentin de Quelen <quentin@dequelen.me>",
    "Clément Renault <clement@meilisearch.com>",
--- a/README.md
+++ b/README.md
@ -25,7 +25,7 @@

 <p align="center">⚡ A lightning-fast search engine that fits effortlessly into your apps, websites, and workflow 🔍</p>

-Meilisearch helps you shape a delightful search experience in a snap, offering features that work out-of-the-box to speed up your workflow.
+[Meilisearch](https://www.meilisearch.com) helps you shape a delightful search experience in a snap, offering features that work out of the box to speed up your workflow.

 <p align="center" name="demo">
  <a href="https://where2watch.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demo-gif#gh-light-mode-only" target="_blank">
@ -39,8 +39,8 @@ Meilisearch helps you shape a delightful search experience in a snap, offering f
 🔥 [**Try it!**](https://where2watch.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demo-link) 🔥

 ## ✨ Features
-
- **Search-as-you-type:** find search results in less than 50 milliseconds
+- **Hybrid search:** Combine the best of both [semantic](https://www.meilisearch.com/docs/learn/experimental/vector_search) & full-text search to get the most relevant results
+- **Search-as-you-type:** find & display results in less than 50 milliseconds to provide an intuitive experience
 - **[Typo tolerance](https://www.meilisearch.com/docs/learn/configuration/typo_tolerance?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** get relevant matches even when queries contain typos and misspellings
 - **[Filtering](https://www.meilisearch.com/docs/learn/fine_tuning_results/filtering?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features) and [faceted search](https://www.meilisearch.com/docs/learn/fine_tuning_results/faceted_search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** enhance your users' search experience with custom filters and build a faceted search interface in a few lines of code
 - **[Sorting](https://www.meilisearch.com/docs/learn/fine_tuning_results/sorting?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=features):** sort results based on price, date, or pretty much anything else your users need
@ -55,15 +55,15 @@ Meilisearch helps you shape a delightful search experience in a snap, offering f

 ## 📖 Documentation

-You can consult Meilisearch's documentation at [https://www.meilisearch.com/docs](https://www.meilisearch.com/docs/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=docs).
+You can consult Meilisearch's documentation at [meilisearch.com/docs](https://www.meilisearch.com/docs/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=docs).

 ## 🚀 Getting started

 For basic instructions on how to set up Meilisearch, add documents to an index, and search for documents, take a look at our [Quick Start](https://www.meilisearch.com/docs/learn/getting_started/quick_start?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=get-started) guide.

-## ⚡ Supercharge your Meilisearch experience
+## 🌍 Supercharge your Meilisearch experience

-Say goodbye to server deployment and manual updates with [Meilisearch Cloud](https://www.meilisearch.com/cloud?utm_campaign=oss&utm_source=github&utm_medium=meilisearch). No credit card required.
+Say goodbye to server deployment and manual updates with [Meilisearch Cloud](https://www.meilisearch.com/cloud?utm_campaign=oss&utm_source=github&utm_medium=meilisearch). Additional features include analytics & monitoring in many regions around the world. No credit card is required.

 ## 🧰 SDKs & integration tools

@ -85,13 +85,13 @@ Finally, for more in-depth information, refer to our articles explaining fundame

 Meilisearch collects **anonymized** data from users to help us improve our product. You can [deactivate this](https://www.meilisearch.com/docs/learn/what_is_meilisearch/telemetry?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=telemetry#how-to-disable-data-collection) whenever you want.

-To request deletion of collected data, please write to us at [privacy@meilisearch.com](mailto:privacy@meilisearch.com). Don't forget to include your `Instance UID` in the message, as this helps us quickly find and delete your data.
+To request deletion of collected data, please write to us at [privacy@meilisearch.com](mailto:privacy@meilisearch.com). Remember to include your `Instance UID` in the message, as this helps us quickly find and delete your data.

 If you want to know more about the kind of data we collect and what we use it for, check the [telemetry section](https://www.meilisearch.com/docs/learn/what_is_meilisearch/telemetry?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=telemetry#how-to-disable-data-collection) of our documentation.

 ## 📫 Get in touch!

-Meilisearch is a search engine created by [Meili](https://www.welcometothejungle.com/en/companies/meilisearch), a software development company based in France and with team members all over the world. Want to know more about us? [Check out our blog!](https://blog.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=contact)
+Meilisearch is a search engine created by [Meili]([https://www.welcometothejungle.com/en/companies/meilisearch](https://www.meilisearch.com/careers)), a software development company headquartered in France and with team members all over the world. Want to know more about us? [Check out our blog!](https://blog.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=contact)

 🗞 [Subscribe to our newsletter](https://meilisearch.us2.list-manage.com/subscribe?u=27870f7b71c908a8b359599fb&id=79582d828e) if you don't want to miss any updates! We promise we won't clutter your mailbox: we only send one edition every two months.

--- a/dump/src/reader/mod.rs
+++ b/dump/src/reader/mod.rs
@ -197,6 +197,140 @@ pub(crate) mod test {
    use super::*;
    use crate::reader::v6::RuntimeTogglableFeatures;

+    #[test]
+    fn import_dump_v6_with_vectors() {
+        // dump containing two indexes
+        //
+        // "vector", configured with an embedder
+        // contains:
+        // - one document with an overriden vector,
+        // - one document with a natural vector
+        // - one document with a _vectors map containing one additional embedder name and a natural vector
+        // - one document with a _vectors map containing one additional embedder name and an overriden vector
+        //
+        // "novector", no embedder
+        // contains:
+        // - a document without vector
+        // - a document with a random _vectors field
+        let dump = File::open("tests/assets/v6-with-vectors.dump").unwrap();
+        let mut dump = DumpReader::open(dump).unwrap();
+
+        // top level infos
+        insta::assert_display_snapshot!(dump.date().unwrap(), @"2024-05-16 15:51:34.151044 +00:00:00");
+        insta::assert_debug_snapshot!(dump.instance_uid().unwrap(), @"None");
+
+        // tasks
+        let tasks = dump.tasks().unwrap().collect::<Result<Vec<_>>>().unwrap();
+        let (tasks, update_files): (Vec<_>, Vec<_>) = tasks.into_iter().unzip();
+        meili_snap::snapshot_hash!(meili_snap::json_string!(tasks), @"278f63325ef06ca04d01df98d8207b94");
+        assert_eq!(update_files.len(), 10);
+        assert!(update_files[0].is_none()); // the dump creation
+        assert!(update_files[1].is_none());
+        assert!(update_files[2].is_none());
+        assert!(update_files[3].is_none());
+        assert!(update_files[4].is_none());
+        assert!(update_files[5].is_none());
+        assert!(update_files[6].is_none());
+        assert!(update_files[7].is_none());
+        assert!(update_files[8].is_none());
+        assert!(update_files[9].is_none());
+
+        // indexes
+        let mut indexes = dump.indexes().unwrap().collect::<Result<Vec<_>>>().unwrap();
+        // the index are not ordered in any way by default
+        indexes.sort_by_key(|index| index.metadata().uid.to_string());
+
+        let mut vector_index = indexes.pop().unwrap();
+        let mut novector_index = indexes.pop().unwrap();
+        assert!(indexes.is_empty());
+
+        // vector
+
+        insta::assert_json_snapshot!(vector_index.metadata(), @r###"
+        {
+          "uid": "vector",
+          "primaryKey": "id",
+          "createdAt": "2024-05-16T15:33:17.240962Z",
+          "updatedAt": "2024-05-16T15:40:55.723052Z"
+        }
+        "###);
+
+        {
+            let documents: Result<Vec<_>> = vector_index.documents().unwrap().collect();
+            let mut documents = documents.unwrap();
+            assert_eq!(documents.len(), 4);
+
+            documents.sort_by_key(|doc| doc.get("id").unwrap().to_string());
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document);
+            }
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document);
+            }
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document);
+            }
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document);
+            }
+        }
+
+        // novector
+
+        insta::assert_json_snapshot!(novector_index.metadata(), @r###"
+        {
+          "uid": "novector",
+          "primaryKey": "id",
+          "createdAt": "2024-05-16T15:33:03.568055Z",
+          "updatedAt": "2024-05-16T15:33:07.530217Z"
+        }
+        "###);
+
+        insta::assert_json_snapshot!(novector_index.settings().unwrap().embedders, @"null");
+
+        {
+            let documents: Result<Vec<_>> = novector_index.documents().unwrap().collect();
+            let mut documents = documents.unwrap();
+            assert_eq!(documents.len(), 2);
+
+            documents.sort_by_key(|doc| doc.get("id").unwrap().to_string());
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document, @r###"
+                {
+                  "id": "e1",
+                  "other": "random1",
+                  "_vectors": "toto"
+                }
+                "###);
+            }
+
+            {
+                let document = documents.pop().unwrap();
+                insta::assert_json_snapshot!(document, @r###"
+                {
+                  "id": "e0",
+                  "other": "random0"
+                }
+                "###);
+            }
+        }
+
+        assert_eq!(
+            dump.features().unwrap().unwrap(),
+            RuntimeTogglableFeatures { vector_store: true, ..Default::default() }
+        );
+    }
+
    #[test]
    fn import_dump_v6_experimental() {
        let dump = File::open("tests/assets/v6-with-experimental.dump").unwrap();
--- a/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-5.snap
+++ b/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-5.snap
@ -0,0 +1,783 @@
+---
+source: dump/src/reader/mod.rs
+expression: document
+---
+{
+  "id": "e3",
+  "desc": "overriden vector + map",
+  "_vectors": {
+    "default": [
+      0.2,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1
+    ],
+    "toto": [
+      0.1
+    ]
+  }
+}
--- a/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-6.snap
+++ b/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-6.snap
@ -0,0 +1,786 @@
+---
+source: dump/src/reader/mod.rs
+expression: document
+---
+{
+  "id": "e2",
+  "desc": "natural vector + map",
+  "_vectors": {
+    "toto": [],
+    "default": {
+      "embeddings": [
+        [
+          -0.05189208313822746,
+          -0.9273212552070618,
+          0.1443813145160675,
+          0.0932632014155388,
+          0.2665371894836426,
+          0.36266782879829407,
+          0.6402910947799683,
+          0.32014018297195435,
+          0.030915971845388412,
+          -0.9312191605567932,
+          -0.3718109726905823,
+          -0.2700554132461548,
+          -1.1014580726623535,
+          0.9154956936836244,
+          -0.3406888246536255,
+          1.0077725648880005,
+          0.6577560901641846,
+          -0.3955195546150207,
+          -0.4148270785808563,
+          0.1855088472366333,
+          0.5062315464019775,
+          -0.3632686734199524,
+          -0.2277890294790268,
+          0.2560805082321167,
+          -0.3853609561920166,
+          -0.1604762226343155,
+          -0.13947471976280212,
+          -0.20147813856601715,
+          -0.4466346800327301,
+          -0.3761846721172333,
+          0.1443382054567337,
+          0.18205296993255615,
+          0.49359792470932007,
+          -0.22538000345230105,
+          -0.4996317625045776,
+          -0.22734887897968292,
+          -0.6034309267997742,
+          -0.7857939600944519,
+          -0.34923747181892395,
+          -0.3466345965862274,
+          0.21176661550998688,
+          -0.5101462006568909,
+          -0.3403083384037018,
+          0.000315118464641273,
+          0.236465722322464,
+          -0.10246097296476364,
+          -1.3013339042663574,
+          0.3419138789176941,
+          -0.32963496446609497,
+          -0.0901619717478752,
+          -0.5426247119903564,
+          0.22656650841236117,
+          -0.44758284091949463,
+          0.14151698350906372,
+          -0.1089438870549202,
+          0.5500766634941101,
+          -0.670711100101471,
+          -0.6227269768714905,
+          0.3894464075565338,
+          -0.27609574794769287,
+          0.7028202414512634,
+          -0.19697771966457367,
+          0.328511506319046,
+          0.5063360929489136,
+          0.4065195322036743,
+          0.2614171802997589,
+          -0.30274391174316406,
+          1.0393824577331543,
+          -0.7742937207221985,
+          -0.7874112129211426,
+          -0.6749666929244995,
+          0.5190866589546204,
+          0.004123548045754433,
+          -0.28312963247299194,
+          -0.038731709122657776,
+          -1.0142987966537476,
+          -0.09519586712121964,
+          0.8755272626876831,
+          0.4876938760280609,
+          0.7811151742935181,
+          0.85174959897995,
+          0.11826585978269576,
+          0.5373436808586121,
+          0.3649002015590668,
+          0.19064077734947205,
+          -0.00287026260048151,
+          -0.7305403351783752,
+          -0.015206154435873032,
+          -0.7899249196052551,
+          0.19407285749912265,
+          0.08596625179052353,
+          -0.28976231813430786,
+          -0.1525907665491104,
+          0.3798313438892365,
+          0.050306469202041626,
+          -0.5697937607765198,
+          0.4219021201133728,
+          0.276252806186676,
+          0.1559903472661972,
+          0.10030482709407806,
+          -0.4043720066547394,
+          -0.1969818025827408,
+          0.5739826560020447,
+          0.2116064727306366,
+          -1.4620544910430908,
+          -0.7802462577819824,
+          -0.24739810824394223,
+          -0.09791352599859238,
+          -0.4413802027702331,
+          0.21549351513385773,
+          -0.9520436525344848,
+          -0.08762510865926743,
+          0.08154498040676117,
+          -0.6154940724372864,
+          -1.01079523563385,
+          0.885427713394165,
+          0.6967288851737976,
+          0.27186504006385803,
+          -0.43194177746772766,
+          -0.11248451471328735,
+          0.7576630711555481,
+          0.4998855590820313,
+          0.0264343973249197,
+          0.9872855544090272,
+          0.5634694695472717,
+          0.053698331117630005,
+          0.19410227239131927,
+          0.3570743501186371,
+          -0.23670297861099243,
+          -0.9114483594894408,
+          0.07884842902421951,
+          0.7318344116210938,
+          0.44630110263824463,
+          0.08745364099740982,
+          -0.347101628780365,
+          -0.4314247667789459,
+          -0.5060274004936218,
+          0.003706763498485088,
+          0.44320008158683777,
+          -0.00788921769708395,
+          -0.1368623524904251,
+          -0.17391923069953918,
+          0.14473655819892883,
+          0.10927865654230118,
+          0.6974599361419678,
+          0.005052129738032818,
+          -0.016953065991401672,
+          -0.1256176233291626,
+          -0.036742497235536575,
+          0.5591985583305359,
+          -0.37619709968566895,
+          0.22429119050502777,
+          0.5403043031692505,
+          -0.8603790998458862,
+          -0.3456307053565979,
+          0.9292937517166138,
+          0.5074859261512756,
+          0.6310645937919617,
+          -0.3091641068458557,
+          0.46902573108673096,
+          0.7891915440559387,
+          0.4499550759792328,
+          0.2744995653629303,
+          0.2712305784225464,
+          -0.04349074140191078,
+          -0.3638863265514374,
+          0.7839881777763367,
+          0.7352104783058167,
+          -0.19457511603832245,
+          -0.5957832932472229,
+          -0.43704694509506226,
+          -1.084769368171692,
+          0.4904985725879669,
+          0.5385226011276245,
+          0.1891629993915558,
+          0.12338479608297348,
+          0.8315675258636475,
+          -0.07830192148685455,
+          1.0916285514831543,
+          -0.28066861629486084,
+          -1.3585069179534912,
+          0.5203898549079895,
+          0.08678033947944641,
+          -0.2566044330596924,
+          0.09484415501356123,
+          -0.0180208683013916,
+          1.0264745950698853,
+          -0.023572135716676712,
+          0.5864979028701782,
+          0.7625196576118469,
+          -0.2543414533138275,
+          -0.8877770900726318,
+          0.7611982822418213,
+          -0.06220436468720436,
+          0.937336564064026,
+          0.2704363465309143,
+          -0.37733694911003113,
+          0.5076137781143188,
+          -0.30641937255859375,
+          0.6252772808074951,
+          -0.0823579877614975,
+          -0.03736555948853493,
+          0.4131673276424408,
+          -0.6514252424240112,
+          0.12918265163898468,
+          -0.4483584463596344,
+          0.6750786304473877,
+          -0.37008383870124817,
+          -0.02324833907186985,
+          0.38027650117874146,
+          -0.26374951004981995,
+          0.4346931278705597,
+          0.42882832884788513,
+          -0.48798441886901855,
+          1.1882442235946655,
+          0.5132288336753845,
+          0.5284568667411804,
+          -0.03538886830210686,
+          0.29620853066444397,
+          -1.0683696269989014,
+          0.25936177372932434,
+          0.10404160618782043,
+          -0.25796034932136536,
+          0.027896970510482788,
+          -0.09225251525640488,
+          1.4811025857925415,
+          0.641173779964447,
+          -0.13838383555412292,
+          -0.3437179923057556,
+          0.5667019486427307,
+          -0.5400741696357727,
+          0.31090837717056274,
+          0.6470608115196228,
+          -0.3747067153453827,
+          -0.7364534735679626,
+          -0.07431528717279434,
+          0.5173454880714417,
+          -0.6578747034072876,
+          0.7107478976249695,
+          -0.7918999791145325,
+          -0.0648345872759819,
+          0.609937846660614,
+          -0.7329513430595398,
+          0.9741371870040894,
+          0.17912346124649048,
+          -0.02658769302070141,
+          0.5162150859832764,
+          -0.3978803157806397,
+          -0.7833885550498962,
+          -0.6497276425361633,
+          -0.3898126780986786,
+          -0.0952848568558693,
+          0.2663288116455078,
+          -0.1604052186012268,
+          0.373076468706131,
+          -0.8357769250869751,
+          -0.05217683315277099,
+          -0.2680160701274872,
+          0.8389158248901367,
+          0.6833611130714417,
+          -0.6712407469749451,
+          0.7406917214393616,
+          -0.44522786140441895,
+          -0.34645363688468933,
+          -0.27384576201438904,
+          -0.9878405928611756,
+          -0.8166060447692871,
+          0.06268279999494553,
+          0.38567957282066345,
+          -0.3274703919887543,
+          0.5296315550804138,
+          -0.11810623109340668,
+          0.23029841482639313,
+          0.08616159111261368,
+          -0.2195747196674347,
+          0.09430307894945145,
+          0.4057176411151886,
+          0.4892159104347229,
+          -0.1636916548013687,
+          -0.6071445345878601,
+          0.41256585717201233,
+          0.622254490852356,
+          -0.41223976016044617,
+          -0.6686707139015198,
+          -0.7474371790885925,
+          -0.8509522080421448,
+          -0.16754287481307983,
+          -0.9078601002693176,
+          -0.29653599858283997,
+          -0.5020652413368225,
+          0.4692700505256653,
+          0.01281109917908907,
+          -0.16071580350399017,
+          0.03388889133930206,
+          -0.020511148497462273,
+          0.5027827024459839,
+          -0.20729811489582065,
+          0.48107290267944336,
+          0.33669769763946533,
+          -0.5275911688804626,
+          0.48271527886390686,
+          0.2738940715789795,
+          -0.033152539283037186,
+          -0.13629786670207977,
+          -0.05965912342071533,
+          -0.26200807094573975,
+          0.04002794995903969,
+          -0.34095603227615356,
+          -3.986898899078369,
+          -0.46819332242012024,
+          -0.422744482755661,
+          -0.169097900390625,
+          0.6008929014205933,
+          0.058016058057546616,
+          -0.11401277780532836,
+          -0.3077819049358368,
+          -0.09595538675785063,
+          0.6723822355270386,
+          0.19367831945419312,
+          0.28304359316825867,
+          0.1609862744808197,
+          0.7567598819732666,
+          0.6889985799789429,
+          0.06907720118761063,
+          -0.04188092052936554,
+          -0.7434936165809631,
+          0.13321782648563385,
+          0.8456063270568848,
+          -0.10364038497209548,
+          -0.45084846019744873,
+          -0.4758241474628449,
+          0.43882066011428833,
+          -0.6432598829269409,
+          0.7217311859130859,
+          -0.24189773201942444,
+          0.12737572193145752,
+          -1.1008601188659668,
+          -0.3305315673351288,
+          0.14614742994308472,
+          -0.7819333076477051,
+          0.5287120342254639,
+          -0.055538054555654526,
+          0.1877404749393463,
+          -0.6907662153244019,
+          0.5616975426673889,
+          -0.4611121714115143,
+          -0.26109233498573303,
+          -0.12898315489292145,
+          -0.3724522292613983,
+          -0.7191406488418579,
+          -0.4425233602523804,
+          -0.644108235836029,
+          0.8424481153488159,
+          0.17532426118850708,
+          -0.5121750235557556,
+          -0.6467239260673523,
+          -0.0008507720194756985,
+          0.7866212129592896,
+          -0.02644744887948036,
+          -0.005045140627771616,
+          0.015782782807946205,
+          0.16334445774555206,
+          -0.1913367658853531,
+          -0.13697923719882965,
+          -0.6684983372688293,
+          0.18346354365348816,
+          -0.341105580329895,
+          0.5427411198616028,
+          0.3779832422733307,
+          -0.6778115034103394,
+          -0.2931850254535675,
+          -0.8805161714553833,
+          -0.4212774932384491,
+          -0.5368952751159668,
+          -1.3937891721725464,
+          -1.225494146347046,
+          0.4276703894138336,
+          1.1205668449401855,
+          -0.6005299687385559,
+          0.15732505917549133,
+          -0.3914784789085388,
+          -1.357046604156494,
+          -0.4707142114639282,
+          -0.1497287154197693,
+          -0.25035548210144043,
+          -0.34328439831733704,
+          0.39083412289619446,
+          0.1623048633337021,
+          -0.9275814294815063,
+          -0.6430015563964844,
+          0.2973862886428833,
+          0.5580436587333679,
+          -0.6232585310935974,
+          -0.6611042022705078,
+          0.4015969038009643,
+          -1.0232892036437988,
+          -0.2585645020008087,
+          -0.5431421399116516,
+          0.5021264553070068,
+          -0.48601630330085754,
+          -0.010242084041237833,
+          0.5862035155296326,
+          0.7316920161247253,
+          0.4036808013916016,
+          0.4269520044326782,
+          -0.705938458442688,
+          0.7747307419776917,
+          0.10164368897676468,
+          0.7887958884239197,
+          -0.9612497091293336,
+          0.12755516171455383,
+          0.06812842190265656,
+          -0.022603651508688927,
+          0.14722754061222076,
+          -0.5588505268096924,
+          -0.20689940452575684,
+          0.3557641804218292,
+          -0.6812759637832642,
+          0.2860803008079529,
+          -0.38954633474349976,
+          0.1759403496980667,
+          -0.5678874850273132,
+          -0.1692986786365509,
+          -0.14578519761562347,
+          0.5711379051208496,
+          1.0208125114440918,
+          0.7759483456611633,
+          -0.372348427772522,
+          -0.5460885763168335,
+          0.7190321683883667,
+          -0.6914990544319153,
+          0.13365162909030914,
+          -0.4854792356491089,
+          0.4054908752441406,
+          0.4502798914909363,
+          -0.3041122555732727,
+          -0.06726965308189392,
+          -0.05570871382951737,
+          -0.0455719493329525,
+          0.4785125255584717,
+          0.8867972493171692,
+          0.4107886850833893,
+          0.6121342182159424,
+          -0.20477132499217987,
+          -0.5598517656326294,
+          -0.6443566679954529,
+          -0.5905212759971619,
+          -0.5571200251579285,
+          0.17573799192905426,
+          -0.28621870279312134,
+          0.1685224026441574,
+          0.09719007462263109,
+          -0.04223639518022537,
+          -0.28623101115226746,
+          -0.1449810117483139,
+          -0.3789580464363098,
+          -0.5227636098861694,
+          -0.049728814512491226,
+          0.7849089503288269,
+          0.16792525351047516,
+          0.9849340915679932,
+          -0.6559549570083618,
+          0.35723909735679626,
+          -0.6822739243507385,
+          1.2873116731643677,
+          0.19993330538272855,
+          0.03512010723352432,
+          -0.6972134113311768,
+          0.18453484773635864,
+          -0.2437680810689926,
+          0.2156416028738022,
+          0.5230382680892944,
+          0.22020135819911957,
+          0.8314080238342285,
+          0.15627102553844452,
+          -0.7330264449119568,
+          0.3888184726238251,
+          -0.22034703195095065,
+          0.5457669496536255,
+          -0.48084837198257446,
+          -0.45576658844947815,
+          -0.09287727624177931,
+          -0.06968110054731369,
+          0.35125672817230225,
+          -0.4278119504451752,
+          0.2038476765155792,
+          0.11392722278833388,
+          0.9433983564376832,
+          -0.4097744226455689,
+          0.035297419875860214,
+          -0.4274404048919678,
+          -0.25100165605545044,
+          1.0943366289138794,
+          -0.07634022831916809,
+          -0.2925529479980469,
+          -0.7512530088424683,
+          0.2649727463722229,
+          -0.4078235328197479,
+          -0.3372223973274231,
+          0.05190162733197212,
+          0.005654910113662481,
+          -0.0001571219472680241,
+          -0.35445958375930786,
+          -0.7837416529655457,
+          0.1500556766986847,
+          0.4383024573326111,
+          0.6099548935890198,
+          0.05951934307813645,
+          -0.21325334906578064,
+          0.0199207104742527,
+          -0.22704418003559113,
+          -0.6481077671051025,
+          0.37442275881767273,
+          -1.015955924987793,
+          0.38637226819992065,
+          -0.06489371508359909,
+          -0.494120329618454,
+          0.3469836115837097,
+          0.15402406454086304,
+          -0.7660972476005554,
+          -0.7053225040435791,
+          -0.25964751839637756,
+          0.014004424214363098,
+          -0.2860170006752014,
+          -0.17565494775772095,
+          -0.45117494463920593,
+          -0.0031954257283359766,
+          0.09676837921142578,
+          -0.514464259147644,
+          0.41698193550109863,
+          -0.21642713248729703,
+          -0.5398141145706177,
+          -0.3647628426551819,
+          0.37005379796028137,
+          0.239425927400589,
+          -0.08833975344896317,
+          0.934946596622467,
+          -0.48340797424316406,
+          0.6241437792778015,
+          -0.7253676652908325,
+          -0.04303571209311485,
+          1.1125205755233765,
+          -0.15692919492721558,
+          -0.2914651036262512,
+          -0.5117168426513672,
+          0.21365483105182648,
+          0.4924402534961701,
+          0.5269662141799927,
+          0.0352792888879776,
+          -0.149167999625206,
+          -0.6019760370254517,
+          0.08245442807674408,
+          0.4900692105293274,
+          0.518824577331543,
+          -0.00005570516441366635,
+          -0.553304135799408,
+          0.22217543423175812,
+          0.5047767758369446,
+          0.135724738240242,
+          1.1511540412902832,
+          -0.3541218340396881,
+          -0.9712511897087096,
+          0.8353699445724487,
+          -0.39227569103240967,
+          -0.9117669463157654,
+          -0.26349931955337524,
+          0.05597023293375969,
+          0.20695461332798004,
+          0.3178807199001312,
+          1.0663238763809204,
+          0.5062212347984314,
+          0.7288597822189331,
+          0.09899299591779707,
+          0.553720235824585,
+          0.675009548664093,
+          -0.20067055523395536,
+          0.3138423264026642,
+          -0.6886593103408813,
+          -0.2910398542881012,
+          -1.3186300992965698,
+          -0.4684459865093231,
+          -0.095743365585804,
+          -0.1257995069026947,
+          -0.4858281314373016,
+          -0.4935407340526581,
+          -0.3266896903514862,
+          -0.3928797245025635,
+          -0.40803104639053345,
+          -0.9975396394729614,
+          0.4229583740234375,
+          0.37309643626213074,
+          0.4431034922599793,
+          0.30364808440208435,
+          -0.3765178918838501,
+          0.5616499185562134,
+          0.16904796659946442,
+          -0.7343707084655762,
+          0.2560209631919861,
+          0.6166825294494629,
+          0.3200829327106476,
+          -0.4483652710914612,
+          0.16224201023578644,
+          -0.31495288014411926,
+          -0.42713335156440735,
+          0.7270734906196594,
+          0.7049484848976135,
+          -0.0571461021900177,
+          0.04477125033736229,
+          -0.6647796034812927,
+          1.183672308921814,
+          0.36199676990509033,
+          0.046881116926670074,
+          0.4515796303749085,
+          0.9278061985969543,
+          0.31471705436706543,
+          -0.7073333859443665,
+          -0.3443860113620758,
+          0.5440067052841187,
+          -0.15020819008350372,
+          -0.541202962398529,
+          0.5203295946121216,
+          1.2192286252975464,
+          -0.9983593225479126,
+          -0.18758884072303772,
+          0.2758221924304962,
+          -0.6511523723602295,
+          -0.1584404855966568,
+          -0.236241415143013,
+          0.2692437767982483,
+          -0.4941152036190033,
+          0.4987454116344452,
+          -0.3331359028816223,
+          0.3163745701313019,
+          0.745529294013977,
+          -0.2905873656272888,
+          0.13602906465530396,
+          0.4679684340953827,
+          1.0555986166000366,
+          1.075700044631958,
+          0.5368486046791077,
+          -0.5118206739425659,
+          0.8668332099914551,
+          -0.5726966857910156,
+          -0.7811751961708069,
+          0.1938626915216446,
+          -0.1929349899291992,
+          0.1757766306400299,
+          0.6384295225143433,
+          0.26462844014167786,
+          0.9542630314826964,
+          0.19313029944896695,
+          1.264248013496399,
+          -0.6304428577423096,
+          0.0487106591463089,
+          -0.16211535036563873,
+          -0.7894763350486755,
+          0.3582514822483063,
+          -0.04153040423989296,
+          0.635784387588501,
+          0.6554391980171204,
+          -0.47010496258735657,
+          -0.8302040696144104,
+          -0.1350124627351761,
+          0.2568812072277069,
+          0.13614831864833832,
+          -0.2563649117946625,
+          -1.0434694290161133,
+          0.3232482671737671,
+          0.47882452607154846,
+          0.4298652410507202,
+          1.0563770532608032,
+          -0.28917592763900757,
+          -0.8533256649971008,
+          0.10648339986801147,
+          0.6376127004623413,
+          -0.20832888782024384,
+          0.2370245456695557,
+          0.0018312990432605147,
+          -0.2034837007522583,
+          0.01051164511591196,
+          -1.105310082435608,
+          0.29724350571632385,
+          0.15604574978351593,
+          0.1973688006401062,
+          0.44394731521606445,
+          0.3974513411521912,
+          -0.13625948131084442,
+          0.9571986198425292,
+          0.2257384955883026,
+          0.2323588728904724,
+          -0.5583669543266296,
+          -0.7854922413825989,
+          0.1647188365459442,
+          -1.6098142862319946,
+          0.318587988615036,
+          -0.13399995863437653,
+          -0.2172701060771942,
+          -0.767514705657959,
+          -0.5813586711883545,
+          -0.3195130527019501,
+          -0.04894036799669266,
+          0.2929930090904236,
+          -0.8213384747505188,
+          0.07181350141763687,
+          0.7469993829727173,
+          0.6407455801963806,
+          0.16365697979927063,
+          0.7870153188705444,
+          0.6524736881256104,
+          0.6399973630905151,
+          -0.04992736503481865,
+          -0.03959266096353531,
+          -0.2512352466583252,
+          0.8448855876922607,
+          -0.1422702670097351,
+          0.1216789186000824,
+          -1.2647287845611572,
+          0.5931149125099182,
+          0.7186052203178406,
+          -0.06118432432413101,
+          -1.1942816972732544,
+          -0.17677085101604462,
+          0.31543800234794617,
+          -0.32252824306488037,
+          0.8255583047866821,
+          -0.14529970288276672,
+          -0.2695446312427521,
+          -0.33378756046295166,
+          -0.1653425395488739,
+          0.1454019844532013,
+          -0.3920115828514099,
+          0.912214994430542,
+          -0.7279734015464783,
+          0.7374742031097412,
+          0.933980405330658,
+          0.13429680466651917,
+          -0.514870285987854,
+          0.3989711999893189,
+          -0.11613689363002776,
+          0.4022413492202759,
+          -0.9990655779838562,
+          -0.33749932050704956,
+          -0.4334589838981629,
+          -1.376373291015625,
+          -0.2993924915790558,
+          -0.09454808384180068,
+          -0.01314175222069025,
+          -0.001090060803107917,
+          0.2137461006641388,
+          0.2938512861728668,
+          0.17508235573768616,
+          0.8260607123374939,
+          -0.7218498587608337,
+          0.2414487451314926,
+          -0.47296759486198425,
+          -0.3002610504627228,
+          -1.238540768623352,
+          0.08663805574178696,
+          0.6805586218833923,
+          0.5909030437469482,
+          -0.42807504534721375,
+          -0.22887496650218964,
+          0.47537800669670105,
+          -1.0474627017974854,
+          0.6338009238243103,
+          0.06548397243022919,
+          0.4971011281013489,
+          1.3484878540039063
+        ]
+      ],
+      "userProvided": false
+    }
+  }
+}
--- a/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-7.snap
+++ b/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-7.snap
@ -0,0 +1,785 @@
+---
+source: dump/src/reader/mod.rs
+expression: document
+---
+{
+  "id": "e1",
+  "desc": "natural vector",
+  "_vectors": {
+    "default": {
+      "embeddings": [
+        [
+          -0.2979458272457123,
+          -0.5288640856742859,
+          -0.019957859069108963,
+          -0.18495318293571472,
+          0.7429973483085632,
+          0.5238497257232666,
+          0.432366281747818,
+          0.32744166254997253,
+          0.0020762972999364138,
+          -0.9507834911346436,
+          -0.35097137093544006,
+          0.08469701558351517,
+          -1.4176613092422483,
+          0.4647577106952667,
+          -0.69340580701828,
+          1.0372896194458008,
+          0.3716741800308227,
+          0.06031008064746857,
+          -0.6152024269104004,
+          0.007914665155112743,
+          0.7954924702644348,
+          -0.20773003995418549,
+          0.09376765787601472,
+          0.04508133605122566,
+          -0.2084471583366394,
+          -0.1518009901046753,
+          0.018195509910583496,
+          -0.07044368237257004,
+          -0.18119366466999057,
+          -0.4480230510234833,
+          0.3822529911994934,
+          0.1911812424659729,
+          0.4674372375011444,
+          0.06963984668254852,
+          -0.09341949224472046,
+          0.005675444379448891,
+          -0.6774799227714539,
+          -0.7066726684570313,
+          -0.39256376028060913,
+          0.04005039855837822,
+          0.2084812968969345,
+          -0.7872875928878784,
+          -0.8205880522727966,
+          0.2919981777667999,
+          -0.06004738807678223,
+          -0.4907574355602264,
+          -1.5937862396240234,
+          0.24249385297298431,
+          -0.14709846675395966,
+          -0.11860740929841997,
+          -0.8299489617347717,
+          0.472964346408844,
+          -0.497518390417099,
+          -0.22205302119255063,
+          -0.4196169078350067,
+          0.32697558403015137,
+          -0.360930860042572,
+          -0.9789686799049376,
+          0.1887447088956833,
+          -0.403737336397171,
+          0.18524253368377688,
+          0.3768732249736786,
+          0.3666233420372009,
+          0.3511938452720642,
+          0.6985810995101929,
+          0.41721710562705994,
+          0.09754953533411026,
+          0.6204307079315186,
+          -1.0762996673583984,
+          -0.06263761967420578,
+          -0.7376511693000793,
+          0.6849768161773682,
+          -0.1745152473449707,
+          -0.40449759364128113,
+          0.20757411420345304,
+          -0.8424443006515503,
+          0.330015629529953,
+          0.3489064872264862,
+          1.0954371690750122,
+          0.8487558960914612,
+          1.1076823472976685,
+          0.61430823802948,
+          0.4155903458595276,
+          0.4111340939998626,
+          0.05753209814429283,
+          -0.06429877132177353,
+          -0.765606164932251,
+          -0.41703930497169495,
+          -0.508820652961731,
+          0.19859947264194489,
+          -0.16607828438282013,
+          -0.28112146258354187,
+          0.11032675206661224,
+          0.38809511065483093,
+          -0.36498191952705383,
+          -0.48671194911003113,
+          0.6755134463310242,
+          0.03958442434668541,
+          0.4478721618652344,
+          -0.10335399955511092,
+          -0.9546685814857484,
+          -0.6087718605995178,
+          0.17498846352100372,
+          0.08320838958024979,
+          -1.4478336572647097,
+          -0.605027437210083,
+          -0.5867993235588074,
+          -0.14711688458919525,
+          -0.5447602272033691,
+          -0.026259321719408035,
+          -0.6997418403625488,
+          -0.07349082082509995,
+          0.10638900846242905,
+          -0.7133527398109436,
+          -0.9396815299987792,
+          1.087092399597168,
+          1.1885089874267578,
+          0.4011896848678589,
+          -0.4089202582836151,
+          -0.10938972979784012,
+          0.6726722121238708,
+          0.24576938152313232,
+          -0.24247920513153076,
+          1.1499971151351929,
+          0.47813335061073303,
+          -0.05331678315997124,
+          0.32338133454322815,
+          0.4870913326740265,
+          -0.23144258558750153,
+          -1.2023426294326782,
+          0.2349330335855484,
+          1.080536961555481,
+          0.29334118962287903,
+          0.391574501991272,
+          -0.15818795561790466,
+          -0.2948290705680847,
+          -0.024689948186278343,
+          0.06602869182825089,
+          0.5937030911445618,
+          -0.047901444137096405,
+          -0.512734591960907,
+          -0.35780075192451477,
+          0.28751692175865173,
+          0.4298716187477112,
+          0.9242428541183472,
+          -0.17208744585514069,
+          0.11515070497989656,
+          -0.0335976779460907,
+          -0.3422986567020416,
+          0.5344581604003906,
+          0.19895796477794647,
+          0.33001241087913513,
+          0.6390730142593384,
+          -0.6074934005737305,
+          -0.2553696632385254,
+          0.9644920229911804,
+          0.2699219584465027,
+          0.6403993368148804,
+          -0.6380003690719604,
+          -0.027310986071825027,
+          0.638815701007843,
+          0.27719101309776306,
+          -0.13553589582443237,
+          0.750195324420929,
+          0.1224869191646576,
+          -0.20613941550254825,
+          0.8444448709487915,
+          0.16200250387191772,
+          -0.24750925600528717,
+          -0.739950954914093,
+          -0.28443849086761475,
+          -1.176282525062561,
+          0.516107976436615,
+          0.3774825632572174,
+          0.10906043648719788,
+          0.07962015271186829,
+          0.7384604215621948,
+          -0.051241904497146606,
+          1.1730090379714966,
+          -0.4828610122203827,
+          -1.404372215270996,
+          0.8811132311820984,
+          -0.3839482367038727,
+          0.022516896948218346,
+          -0.0491158664226532,
+          -0.43027013540267944,
+          1.2049334049224854,
+          -0.27309560775756836,
+          0.6883630752563477,
+          0.8264574408531189,
+          -0.5020735263824463,
+          -0.4874092042446137,
+          0.6007202863693237,
+          -0.4965405762195587,
+          1.1302915811538696,
+          0.032572727650403976,
+          -0.3731859028339386,
+          0.658271849155426,
+          -0.9023059010505676,
+          0.7400162220001221,
+          0.014550759457051754,
+          -0.19699542224407196,
+          0.2319706380367279,
+          -0.789058268070221,
+          -0.14905710518360138,
+          -0.5826214551925659,
+          0.207652747631073,
+          -0.4507439732551574,
+          -0.3163885474205017,
+          0.3604124188423157,
+          -0.45119962096214294,
+          0.3428427278995514,
+          0.3005594313144684,
+          -0.36026081442832947,
+          1.1014249324798584,
+          0.40884315967559814,
+          0.34991952776908875,
+          -0.1806638240814209,
+          0.27440476417541504,
+          -0.7118373513221741,
+          0.4645499587059021,
+          0.214790478348732,
+          -0.2343102991580963,
+          0.10500429570674896,
+          -0.28034430742263794,
+          1.2267805337905884,
+          1.0561333894729614,
+          -0.497364342212677,
+          -0.6143305897712708,
+          0.24963727593421936,
+          -0.33136463165283203,
+          -0.01473914459347725,
+          0.495918869972229,
+          -0.6985538005828857,
+          -1.0033197402954102,
+          0.35937801003456116,
+          0.6325868368148804,
+          -0.6808838844299316,
+          1.0354058742523191,
+          -0.7214401960372925,
+          -0.33318862318992615,
+          0.874398410320282,
+          -0.6594992280006409,
+          0.6830640435218811,
+          -0.18534131348133087,
+          0.024834271520376205,
+          0.19901277124881744,
+          -0.5992477536201477,
+          -1.2126628160476685,
+          -0.9245557188987732,
+          -0.3898217976093292,
+          -0.1286519467830658,
+          0.4217943847179413,
+          -0.1143646091222763,
+          0.5630772709846497,
+          -0.5240639448165894,
+          0.21152715384960177,
+          -0.3792001008987427,
+          0.8266305327415466,
+          1.170984387397766,
+          -0.8072142004966736,
+          0.11382893472909927,
+          -0.17953898012638092,
+          -0.1789460331201553,
+          -0.15078622102737427,
+          -1.2082908153533936,
+          -0.7812382578849792,
+          -0.10903695970773696,
+          0.7303897142410278,
+          -0.39054441452026367,
+          0.19511254131793976,
+          -0.09121843427419662,
+          0.22400228679180145,
+          0.30143046379089355,
+          0.1141919493675232,
+          0.48112115263938904,
+          0.7307931780815125,
+          0.09701362252235413,
+          -0.2795647978782654,
+          -0.3997688889503479,
+          0.5540812611579895,
+          0.564578115940094,
+          -0.40065160393714905,
+          -0.3629159033298493,
+          -0.3789091110229492,
+          -0.7298538088798523,
+          -0.6996853351593018,
+          -0.4477842152118683,
+          -0.289089560508728,
+          -0.6430277824401855,
+          0.2344944179058075,
+          0.3742927014827728,
+          -0.5079357028007507,
+          0.28841453790664673,
+          0.06515737622976303,
+          0.707315981388092,
+          0.09498685598373412,
+          0.8365515470504761,
+          0.10002726316452026,
+          -0.7695478200912476,
+          0.6264724135398865,
+          0.7562043070793152,
+          -0.23112858831882477,
+          -0.2871039807796478,
+          -0.25010058283805847,
+          0.2783474028110504,
+          -0.03224996477365494,
+          -0.9119359850883484,
+          -3.6940200328826904,
+          -0.5099936127662659,
+          -0.1604711413383484,
+          0.17453284561634064,
+          0.41759559512138367,
+          0.1419190913438797,
+          -0.11362407356500626,
+          -0.33312007784843445,
+          0.11511333286762238,
+          0.4667884409427643,
+          -0.0031647447030991316,
+          0.15879854559898376,
+          0.3042248487472534,
+          0.5404849052429199,
+          0.8515422344207764,
+          0.06286454200744629,
+          0.43790125846862793,
+          -0.8682025074958801,
+          -0.06363756954669952,
+          0.5547921657562256,
+          -0.01483887154608965,
+          -0.07361344993114471,
+          -0.929947018623352,
+          0.3502565622329712,
+          -0.5080993175506592,
+          1.0380364656448364,
+          -0.2017953395843506,
+          0.21319580078125,
+          -1.0763001441955566,
+          -0.556368887424469,
+          0.1949922740459442,
+          -0.6445739269256592,
+          0.6791343688964844,
+          0.21188358962535855,
+          0.3736183941364288,
+          -0.21800459921360016,
+          0.7597446441650391,
+          -0.3732394874095917,
+          -0.4710160195827484,
+          0.025146087631583217,
+          0.05341297015547752,
+          -0.9522109627723694,
+          -0.6000866889953613,
+          -0.08469046652317047,
+          0.5966026186943054,
+          0.3444081246852875,
+          -0.461188405752182,
+          -0.5279349088668823,
+          0.10296865552663804,
+          0.5175143480300903,
+          -0.20671147108078003,
+          0.13392412662506104,
+          0.4812754988670349,
+          0.2993808686733246,
+          -0.3005635440349579,
+          0.5141698122024536,
+          -0.6239235401153564,
+          0.2877119481563568,
+          -0.4452739953994751,
+          0.5621107816696167,
+          0.5047508478164673,
+          -0.4226335883140564,
+          -0.18578553199768064,
+          -1.1967322826385498,
+          0.28178197145462036,
+          -0.8692031502723694,
+          -1.1812998056411743,
+          -1.4526212215423584,
+          0.4645712077617645,
+          0.9327932000160216,
+          -0.6560136675834656,
+          0.461549699306488,
+          -0.5621527433395386,
+          -1.328449010848999,
+          -0.08676894754171371,
+          0.00021918353741057217,
+          -0.18864136934280396,
+          0.1259666532278061,
+          0.18240638077259064,
+          -0.14919660985469818,
+          -0.8965857625007629,
+          -0.7539900541305542,
+          0.013973715715110302,
+          0.504276692867279,
+          -0.704748272895813,
+          -0.6428424119949341,
+          0.6303996443748474,
+          -0.5404738187789917,
+          -0.31176653504371643,
+          -0.21262824535369873,
+          0.18736739456653595,
+          -0.7998970746994019,
+          0.039946746081113815,
+          0.7390344738960266,
+          0.4283199906349182,
+          0.3795057237148285,
+          0.07204607129096985,
+          -0.9230587482452391,
+          0.9440426230430604,
+          0.26272690296173096,
+          0.5598306655883789,
+          -1.0520871877670288,
+          -0.2677186131477356,
+          -0.1888762265443802,
+          0.30426350235939026,
+          0.4746131896972656,
+          -0.5746733546257019,
+          -0.4197768568992615,
+          0.8565112948417664,
+          -0.6767723560333252,
+          0.23448683321475983,
+          -0.2010004222393036,
+          0.4112907350063324,
+          -0.6497949957847595,
+          -0.418667733669281,
+          -0.4950824975967407,
+          0.44438859820365906,
+          1.026281714439392,
+          0.482397586107254,
+          -0.26220494508743286,
+          -0.3640787005424499,
+          0.5907743573188782,
+          -0.8771642446517944,
+          0.09708411991596222,
+          -0.3671700060367584,
+          0.4331349730491638,
+          0.619417667388916,
+          -0.2684665620326996,
+          -0.5123821496963501,
+          -0.1502324342727661,
+          -0.012190685607492924,
+          0.3580845892429352,
+          0.8617186546325684,
+          0.3493645489215851,
+          1.0270192623138428,
+          0.18297909200191495,
+          -0.5881339311599731,
+          -0.1733516901731491,
+          -0.5040576457977295,
+          -0.340370237827301,
+          -0.26767754554748535,
+          -0.28570041060447693,
+          -0.032928116619586945,
+          0.6029254794120789,
+          0.17397655546665192,
+          0.09346921741962431,
+          0.27815181016921997,
+          -0.46699589490890503,
+          -0.8148876428604126,
+          -0.3964351713657379,
+          0.3812595009803772,
+          0.13547226786613464,
+          0.7126688361167908,
+          -0.3473474085330963,
+          -0.06573959439992905,
+          -0.6483767032623291,
+          1.4808889627456665,
+          0.30924928188323975,
+          -0.5085946917533875,
+          -0.8613000512123108,
+          0.3048902451992035,
+          -0.4241599142551422,
+          0.15909206867218018,
+          0.5764641761779785,
+          -0.07879110425710678,
+          1.015336513519287,
+          0.07599356025457382,
+          -0.7025855779647827,
+          0.30047643184661865,
+          -0.35094937682151794,
+          0.2522146999835968,
+          -0.2338722199201584,
+          -0.8326804637908936,
+          -0.13695412874221802,
+          -0.03452421352267265,
+          0.47974953055381775,
+          -0.18385636806488037,
+          0.32438594102859497,
+          0.1797013282775879,
+          0.787494957447052,
+          -0.12579888105392456,
+          -0.07507286965847015,
+          -0.4389670491218567,
+          0.2720070779323578,
+          0.8138866424560547,
+          0.01974171027541161,
+          -0.3057698905467987,
+          -0.6709924936294556,
+          0.0885881632566452,
+          -0.2862754464149475,
+          0.03475658595561981,
+          -0.1285519152879715,
+          0.3838353455066681,
+          -0.2944154739379883,
+          -0.4204859137535095,
+          -0.4416137933731079,
+          0.13426260650157928,
+          0.36733248829841614,
+          0.573428750038147,
+          -0.14928072690963745,
+          -0.026076916605234143,
+          0.33286052942276,
+          -0.5340145826339722,
+          -0.17279052734375,
+          -0.01154550164937973,
+          -0.6620771884918213,
+          0.18390542268753052,
+          -0.08265615254640579,
+          -0.2489682286977768,
+          0.2429984211921692,
+          -0.044153645634651184,
+          -0.986578404903412,
+          -0.33574509620666504,
+          -0.5387663841247559,
+          0.19767941534519196,
+          0.12540718913078308,
+          -0.3403128981590271,
+          -0.4154576361179352,
+          0.17275673151016235,
+          0.09407442808151244,
+          -0.5414086580276489,
+          0.4393929839134216,
+          0.1725579798221588,
+          -0.4998118281364441,
+          -0.6926208138465881,
+          0.16552448272705078,
+          0.6659538149833679,
+          -0.10949844866991044,
+          0.986426830291748,
+          0.01748848147690296,
+          0.4003709554672241,
+          -0.5430638194084167,
+          0.35347291827201843,
+          0.6887399554252625,
+          0.08274628221988678,
+          0.13407137989997864,
+          -0.591465950012207,
+          0.3446292281150818,
+          0.6069018244743347,
+          0.1935492902994156,
+          -0.0989871397614479,
+          0.07008486241102219,
+          -0.8503749370574951,
+          -0.09507356584072112,
+          0.6259510517120361,
+          0.13934025168418884,
+          0.06392545253038406,
+          -0.4112265408039093,
+          -0.08475656062364578,
+          0.4974113404750824,
+          -0.30606114864349365,
+          1.111435890197754,
+          -0.018766529858112335,
+          -0.8422622680664063,
+          0.4325508773326874,
+          -0.2832120656967163,
+          -0.4859798848628998,
+          -0.41498348116874695,
+          0.015977520495653152,
+          0.5292825698852539,
+          0.4538311660289765,
+          1.1328668594360352,
+          0.22632671892642975,
+          0.7918671369552612,
+          0.33401933312416077,
+          0.7306135296821594,
+          0.3548600673675537,
+          0.12506209313869476,
+          0.8573207855224609,
+          -0.5818327069282532,
+          -0.6953738927841187,
+          -1.6171947717666626,
+          -0.1699674427509308,
+          0.6318262815475464,
+          -0.05671752244234085,
+          -0.28145185112953186,
+          -0.3976689279079437,
+          -0.2041076272726059,
+          -0.5495951175689697,
+          -0.5152917504310608,
+          -0.9309796094894408,
+          0.101932130753994,
+          0.1367802917957306,
+          0.1490798443555832,
+          0.5304336547851563,
+          -0.5082434415817261,
+          0.06688683480024338,
+          0.14657628536224365,
+          -0.782435953617096,
+          0.2962816655635834,
+          0.6965363621711731,
+          0.8496337532997131,
+          -0.3042965829372406,
+          0.04343798756599426,
+          0.0330701619386673,
+          -0.5662598013877869,
+          1.1086925268173218,
+          0.756072998046875,
+          -0.204134538769722,
+          0.2404300570487976,
+          -0.47848284244537354,
+          1.3659011125564575,
+          0.5645433068275452,
+          -0.15836156904697418,
+          0.43395575881004333,
+          0.5944653749465942,
+          1.0043466091156006,
+          -0.49446743726730347,
+          -0.5954391360282898,
+          0.5341240763664246,
+          0.020598189905285835,
+          -0.4036853015422821,
+          0.4473709762096405,
+          1.1998231410980225,
+          -0.9317775368690492,
+          -0.23321466147899628,
+          0.2052552700042725,
+          -0.7423108816146851,
+          -0.19917210936546328,
+          -0.1722569614648819,
+          -0.034072667360305786,
+          -0.00671181408688426,
+          0.46396249532699585,
+          -0.1372445821762085,
+          0.053376372903585434,
+          0.7392690777778625,
+          -0.38447609543800354,
+          0.07497968524694443,
+          0.5197252631187439,
+          1.3746477365493774,
+          0.9060075879096984,
+          0.20000585913658145,
+          -0.4053704142570496,
+          0.7497360110282898,
+          -0.34087055921554565,
+          -1.101803183555603,
+          0.273650586605072,
+          -0.5125769376754761,
+          0.22472351789474487,
+          0.480757474899292,
+          -0.19845178723335263,
+          0.8857700824737549,
+          0.30752456188201904,
+          1.1109285354614258,
+          -0.6768012642860413,
+          0.524367094039917,
+          -0.22495046257972717,
+          -0.4224412739276886,
+          0.40753406286239624,
+          -0.23133376240730288,
+          0.3297771215438843,
+          0.4905449151992798,
+          -0.6813114285469055,
+          -0.7543983459472656,
+          -0.5599071383476257,
+          0.14351597428321838,
+          -0.029278717935085297,
+          -0.3970443606376648,
+          -0.303079217672348,
+          0.24161772429943085,
+          0.008353390730917454,
+          -0.0062365154735744,
+          1.0824860334396362,
+          -0.3704061508178711,
+          -1.0337258577346802,
+          0.04638749733567238,
+          1.163011074066162,
+          -0.31737643480300903,
+          0.013986887410283089,
+          0.19223114848136905,
+          -0.2260770797729492,
+          -0.210910826921463,
+          -1.0191949605941772,
+          0.22356095910072327,
+          0.09353553503751756,
+          0.18096882104873657,
+          0.14867214858531952,
+          0.43408671021461487,
+          -0.33312076330184937,
+          0.8173948526382446,
+          0.6428242921829224,
+          0.20215003192424777,
+          -0.6634518504142761,
+          -0.4132290482521057,
+          0.29815030097961426,
+          -1.579406976699829,
+          -0.0981958732008934,
+          -0.03941014781594277,
+          0.1709178239107132,
+          -0.5481140613555908,
+          -0.5338194966316223,
+          -0.3528362512588501,
+          -0.11561278253793716,
+          -0.21793591976165771,
+          -1.1570470333099363,
+          0.2157980799674988,
+          0.42083489894866943,
+          0.9639263153076172,
+          0.09747201204299928,
+          0.15671424567699432,
+          0.4034591615200043,
+          0.6728067994117737,
+          -0.5216875672340393,
+          0.09657668322324751,
+          -0.2416689097881317,
+          0.747975766658783,
+          0.1021689772605896,
+          0.11652665585279463,
+          -1.0484966039657593,
+          0.8489304780960083,
+          0.7169828414916992,
+          -0.09012343734502792,
+          -1.3173753023147583,
+          0.057890523225069046,
+          -0.006231260951608419,
+          -0.1018214002251625,
+          0.936040461063385,
+          -0.0502331368625164,
+          -0.4284322261810303,
+          -0.38209280371665955,
+          -0.22668412327766416,
+          0.0782942995429039,
+          -0.4881664514541626,
+          0.9268959760665894,
+          0.001867273123934865,
+          0.42261114716529846,
+          0.8283362984657288,
+          0.4256294071674347,
+          -0.7965338826179504,
+          0.4840078353881836,
+          -0.19861412048339844,
+          0.33977967500686646,
+          -0.4604192078113556,
+          -0.3107339143753052,
+          -0.2839638590812683,
+          -1.5734281539916992,
+          0.005220232997089624,
+          0.09239906817674635,
+          -0.7828494906425476,
+          -0.1397123783826828,
+          0.2576255202293396,
+          0.21372435986995697,
+          -0.23169949650764465,
+          0.4016408920288086,
+          -0.462497353553772,
+          -0.2186472862958908,
+          -0.5617868900299072,
+          -0.3649831712245941,
+          -1.1585862636566162,
+          -0.08222806453704834,
+          0.931126832962036,
+          0.4327389597892761,
+          -0.46451422572135925,
+          -0.5430706143379211,
+          -0.27434298396110535,
+          -0.9479129314422609,
+          0.1845661848783493,
+          0.3972720205783844,
+          0.4883299469947815,
+          1.04031240940094
+        ]
+      ],
+      "userProvided": false
+    }
+  }
+}
--- a/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-8.snap
+++ b/dump/src/reader/snapshots/dumpreadertest__import_dump_v6_with_vectors-8.snap
@ -0,0 +1,780 @@
+---
+source: dump/src/reader/mod.rs
+expression: document
+---
+{
+  "id": "e0",
+  "desc": "overriden vector",
+  "_vectors": {
+    "default": [
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1,
+      0.1
+    ]
+  }
+}
--- a/dump/tests/assets/v6-with-vectors.dump
+++ b/dump/tests/assets/v6-with-vectors.dump
--- a/index-scheduler/Cargo.toml
+++ b/index-scheduler/Cargo.toml
@ -22,7 +22,6 @@ flate2 = "1.0.28"
 meilisearch-auth = { path = "../meilisearch-auth" }
 meilisearch-types = { path = "../meilisearch-types" }
 page_size = "0.5.0"
-puffin = { version = "0.16.0", features = ["serialization"] }
 rayon = "1.8.1"
 roaring = { version = "0.10.2", features = ["serde"] }
 serde = { version = "1.0.195", features = ["derive"] }
--- a/index-scheduler/src/batch.rs
+++ b/index-scheduler/src/batch.rs
@ -31,6 +31,9 @@ use meilisearch_types::milli::heed::CompactionOption;
 use meilisearch_types::milli::update::{
    IndexDocumentsConfig, IndexDocumentsMethod, IndexerConfig, Settings as MilliSettings,
 };
+use meilisearch_types::milli::vector::parsed_vectors::{
+    ExplicitVectors, VectorOrArrayOfVectors, RESERVED_VECTORS_FIELD_NAME,
+};
 use meilisearch_types::milli::{self, Filter};
 use meilisearch_types::settings::{apply_settings_to_builder, Settings, Unchecked};
 use meilisearch_types::tasks::{Details, IndexSwap, Kind, KindWithContent, Status, Task};
@ -526,8 +529,6 @@ impl IndexScheduler {
        #[cfg(test)]
        self.maybe_fail(crate::tests::FailureLocation::InsideCreateBatch)?;

-        puffin::profile_function!();
-
        let enqueued = &self.get_status(rtxn, Status::Enqueued)?;
        let to_cancel = self.get_kind(rtxn, Kind::TaskCancelation)? & enqueued;

@ -636,8 +637,6 @@ impl IndexScheduler {
            self.breakpoint(crate::Breakpoint::InsideProcessBatch);
        }

-        puffin::profile_function!(batch.to_string());
-
        match batch {
            Batch::TaskCancelation { mut task, previous_started_at, previous_processing_tasks } => {
                // 1. Retrieve the tasks that matched the query at enqueue-time.
@ -785,10 +784,12 @@ impl IndexScheduler {
                let dst = temp_snapshot_dir.path().join("auth");
                fs::create_dir_all(&dst)?;
                // TODO We can't use the open_auth_store_env function here but we should
-                let auth = milli::heed::EnvOpenOptions::new()
-                    .map_size(1024 * 1024 * 1024) // 1 GiB
-                    .max_dbs(2)
-                    .open(&self.auth_path)?;
+                let auth = unsafe {
+                    milli::heed::EnvOpenOptions::new()
+                        .map_size(1024 * 1024 * 1024) // 1 GiB
+                        .max_dbs(2)
+                        .open(&self.auth_path)
+                }?;
                auth.copy_to_file(dst.join("data.mdb"), CompactionOption::Enabled)?;

                // 5. Copy and tarball the flat snapshot
@ -914,8 +915,55 @@ impl IndexScheduler {
                        if self.must_stop_processing.get() {
                            return Err(Error::AbortedTask);
                        }
-                        let (_id, doc) = ret?;
-                        let document = milli::obkv_to_json(&all_fields, &fields_ids_map, doc)?;
+
+                        let (id, doc) = ret?;
+
+                        let mut document = milli::obkv_to_json(&all_fields, &fields_ids_map, doc)?;
+
+                        'inject_vectors: {
+                            let embeddings = index.embeddings(&rtxn, id)?;
+
+                            if embeddings.is_empty() {
+                                break 'inject_vectors;
+                            }
+
+                            let vectors = document
+                                .entry(RESERVED_VECTORS_FIELD_NAME.to_owned())
+                                .or_insert(serde_json::Value::Object(Default::default()));
+
+                            let serde_json::Value::Object(vectors) = vectors else {
+                                return Err(milli::Error::UserError(
+                                    milli::UserError::InvalidVectorsMapType {
+                                        document_id: {
+                                            if let Ok(Some(Ok(index))) = index
+                                                .external_id_of(&rtxn, std::iter::once(id))
+                                                .map(|it| it.into_iter().next())
+                                            {
+                                                index
+                                            } else {
+                                                format!("internal docid={id}")
+                                            }
+                                        },
+                                        value: vectors.clone(),
+                                    },
+                                )
+                                .into());
+                            };
+
+                            for (embedder_name, embeddings) in embeddings {
+                                // don't change the entry if it already exists, because it was user-provided
+                                vectors.entry(embedder_name).or_insert_with(|| {
+                                    let embeddings = ExplicitVectors {
+                                        embeddings: VectorOrArrayOfVectors::from_array_of_vectors(
+                                            embeddings,
+                                        ),
+                                        user_provided: false,
+                                    };
+                                    serde_json::to_value(embeddings).unwrap()
+                                });
+                            }
+                        }
+
                        index_dumper.push_document(&document)?;
                    }

@ -1174,8 +1222,6 @@ impl IndexScheduler {
        index: &'i Index,
        operation: IndexOperation,
    ) -> Result<Vec<Task>> {
-        puffin::profile_function!();
-
        match operation {
            IndexOperation::DocumentClear { mut tasks, .. } => {
                let count = milli::update::ClearDocuments::new(index_wtxn, index).execute()?;
--- a/index-scheduler/src/features.rs
+++ b/index-scheduler/src/features.rs
@ -68,19 +68,6 @@ impl RoFeatures {
            .into())
        }
    }
-
-    pub fn check_puffin(&self) -> Result<()> {
-        if self.runtime.export_puffin_reports {
-            Ok(())
-        } else {
-            Err(FeatureNotEnabledError {
-                disabled_action: "Outputting Puffin reports to disk",
-                feature: "export puffin reports",
-                issue_link: "https://github.com/meilisearch/product/discussions/693",
-            }
-            .into())
-        }
-    }
 }

 impl FeatureData {
--- a/index-scheduler/src/insta_snapshot.rs
+++ b/index-scheduler/src/insta_snapshot.rs
@ -32,7 +32,6 @@ pub fn snapshot_index_scheduler(scheduler: &IndexScheduler) -> String {
        features: _,
        max_number_of_tasks: _,
        max_number_of_batched_tasks: _,
-        puffin_frame: _,
        wake_up: _,
        dumps_path: _,
        snapshots_path: _,
--- a/index-scheduler/src/lib.rs
+++ b/index-scheduler/src/lib.rs
@ -33,7 +33,6 @@ pub type Result<T> = std::result::Result<T, Error>;
 pub type TaskId = u32;

 use std::collections::{BTreeMap, HashMap};
-use std::fs::File;
 use std::io::{self, BufReader, Read};
 use std::ops::{Bound, RangeBounds};
 use std::path::{Path, PathBuf};
@ -59,7 +58,6 @@ use meilisearch_types::milli::vector::{Embedder, EmbedderOptions, EmbeddingConfi
 use meilisearch_types::milli::{self, CboRoaringBitmapCodec, Index, RoaringBitmapCodec, BEU32};
 use meilisearch_types::task_view::TaskView;
 use meilisearch_types::tasks::{Kind, KindWithContent, Status, Task};
-use puffin::FrameView;
 use rayon::current_num_threads;
 use rayon::prelude::{IntoParallelIterator, ParallelIterator};
 use roaring::RoaringBitmap;
@ -344,9 +342,6 @@ pub struct IndexScheduler {
    /// The Authorization header to send to the webhook URL.
    pub(crate) webhook_authorization_header: Option<String>,

-    /// A frame to output the indexation profiling files to disk.
-    pub(crate) puffin_frame: Arc<puffin::GlobalFrameView>,
-
    /// The path used to create the dumps.
    pub(crate) dumps_path: PathBuf,

@ -401,7 +396,6 @@ impl IndexScheduler {
            cleanup_enabled: self.cleanup_enabled,
            max_number_of_tasks: self.max_number_of_tasks,
            max_number_of_batched_tasks: self.max_number_of_batched_tasks,
-            puffin_frame: self.puffin_frame.clone(),
            snapshots_path: self.snapshots_path.clone(),
            dumps_path: self.dumps_path.clone(),
            auth_path: self.auth_path.clone(),
@ -453,10 +447,12 @@ impl IndexScheduler {
            )
        };

-        let env = heed::EnvOpenOptions::new()
-            .max_dbs(11)
-            .map_size(budget.task_db_size)
-            .open(options.tasks_path)?;
+        let env = unsafe {
+            heed::EnvOpenOptions::new()
+                .max_dbs(11)
+                .map_size(budget.task_db_size)
+                .open(options.tasks_path)
+        }?;

        let features = features::FeatureData::new(&env, options.instance_features)?;

@ -498,7 +494,6 @@ impl IndexScheduler {
            env,
            // we want to start the loop right away in case meilisearch was ctrl+Ced while processing things
            wake_up: Arc::new(SignalEvent::auto(true)),
-            puffin_frame: Arc::new(puffin::GlobalFrameView::default()),
            autobatching_enabled: options.autobatching_enabled,
            cleanup_enabled: options.cleanup_enabled,
            max_number_of_tasks: options.max_number_of_tasks,
@ -585,9 +580,9 @@ impl IndexScheduler {
    }

    fn is_good_heed(tasks_path: &Path, map_size: usize) -> bool {
-        if let Ok(env) =
+        if let Ok(env) = unsafe {
            heed::EnvOpenOptions::new().map_size(clamp_to_page_size(map_size)).open(tasks_path)
-        {
+        } {
            env.prepare_for_closing().wait();
            true
        } else {
@ -619,10 +614,6 @@ impl IndexScheduler {
                run.wake_up.wait();

                loop {
-                    let puffin_enabled = run.features().check_puffin().is_ok();
-                    puffin::set_scopes_on(puffin_enabled);
-                    puffin::GlobalProfiler::lock().new_frame();
-
                    match run.tick() {
                        Ok(TickOutcome::TickAgain(_)) => (),
                        Ok(TickOutcome::WaitForSignal) => run.wake_up.wait(),
@ -634,31 +625,6 @@ impl IndexScheduler {
                            }
                        }
                    }
-
-                    // Let's write the previous frame to disk but only if
-                    // the user wanted to profile with puffin.
-                    if puffin_enabled {
-                        let mut frame_view = run.puffin_frame.lock();
-                        if !frame_view.is_empty() {
-                            let now = OffsetDateTime::now_utc();
-                            let mut file = match File::create(format!("{}.puffin", now)) {
-                                Ok(file) => file,
-                                Err(e) => {
-                                    tracing::error!("{e}");
-                                    continue;
-                                }
-                            };
-                            if let Err(e) = frame_view.save_to_writer(&mut file) {
-                                tracing::error!("{e}");
-                            }
-                            if let Err(e) = file.sync_all() {
-                                tracing::error!("{e}");
-                            }
-                            // We erase this frame view as it is no more useful. We want to
-                            // measure the new frames now that we exported the previous ones.
-                            *frame_view = FrameView::default();
-                        }
-                    }
                }
            })
            .unwrap();
@ -1772,6 +1738,7 @@ mod tests {
    use big_s::S;
    use crossbeam::channel::RecvTimeoutError;
    use file_store::File;
+    use insta::assert_json_snapshot;
    use meili_snap::{json_string, snapshot};
    use meilisearch_auth::AuthFilter;
    use meilisearch_types::document_formats::DocumentFormatError;
@ -1849,7 +1816,7 @@ mod tests {

            // To be 100% consistent between all test we're going to start the scheduler right now
            // and ensure it's in the expected starting state.
-            let breakpoint = match receiver.recv_timeout(std::time::Duration::from_secs(1)) {
+            let breakpoint = match receiver.recv_timeout(std::time::Duration::from_secs(10)) {
                Ok(b) => b,
                Err(RecvTimeoutError::Timeout) => {
                    panic!("The scheduler seems to be waiting for a new task while your test is waiting for a breakpoint.")
@ -1960,7 +1927,7 @@ mod tests {
        fn advance(&mut self) -> Breakpoint {
            let (breakpoint_1, b) = match self
                .test_breakpoint_rcv
-                .recv_timeout(std::time::Duration::from_secs(5))
+                .recv_timeout(std::time::Duration::from_secs(50))
            {
                Ok(b) => b,
                Err(RecvTimeoutError::Timeout) => {
@ -1981,7 +1948,7 @@ mod tests {

            let (breakpoint_2, b) = match self
                .test_breakpoint_rcv
-                .recv_timeout(std::time::Duration::from_secs(5))
+                .recv_timeout(std::time::Duration::from_secs(50))
            {
                Ok(b) => b,
                Err(RecvTimeoutError::Timeout) => {
@ -4980,4 +4947,233 @@ mod tests {
        ----------------------------------------------------------------------
        "###);
    }
+
+    #[test]
+    fn import_vectors() {
+        use meilisearch_types::settings::{Settings, Unchecked};
+        use milli::update::Setting;
+
+        let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);
+
+        let mut new_settings: Box<Settings<Unchecked>> = Box::default();
+        let mut embedders = BTreeMap::default();
+        let embedding_settings = milli::vector::settings::EmbeddingSettings {
+            source: Setting::Set(milli::vector::settings::EmbedderSource::Rest),
+            api_key: Setting::Set(S("My super secret")),
+            url: Setting::Set(S("http://localhost:7777")),
+            dimensions: Setting::Set(384),
+            ..Default::default()
+        };
+        embedders.insert(S("A_fakerest"), Setting::Set(embedding_settings));
+
+        let embedding_settings = milli::vector::settings::EmbeddingSettings {
+            source: Setting::Set(milli::vector::settings::EmbedderSource::HuggingFace),
+            model: Setting::Set(S("sentence-transformers/all-MiniLM-L6-v2")),
+            revision: Setting::Set(S("e4ce9877abf3edfe10b0d82785e83bdcb973e22e")),
+            document_template: Setting::Set(S("{{doc.doggo}} the {{doc.breed}} best doggo")),
+            ..Default::default()
+        };
+        embedders.insert(S("B_small_hf"), Setting::Set(embedding_settings));
+
+        new_settings.embedders = Setting::Set(embedders);
+
+        index_scheduler
+            .register(
+                KindWithContent::SettingsUpdate {
+                    index_uid: S("doggos"),
+                    new_settings,
+                    is_deletion: false,
+                    allow_index_creation: true,
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        index_scheduler.assert_internally_consistent();
+
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_registering_settings_task_vectors");
+
+        {
+            let rtxn = index_scheduler.read_txn().unwrap();
+            let task = index_scheduler.get_task(&rtxn, 0).unwrap().unwrap();
+            let task = meilisearch_types::task_view::TaskView::from_task(&task);
+            insta::assert_json_snapshot!(task.details);
+        }
+
+        handle.advance_n_successful_batches(1);
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "settings_update_processed_vectors");
+
+        {
+            let rtxn = index_scheduler.read_txn().unwrap();
+            let task = index_scheduler.get_task(&rtxn, 0).unwrap().unwrap();
+            let task = meilisearch_types::task_view::TaskView::from_task(&task);
+            insta::assert_json_snapshot!(task.details);
+        }
+
+        let (fakerest_name, simple_hf_name, beagle_embed, lab_embed, patou_embed) = {
+            let index = index_scheduler.index("doggos").unwrap();
+            let rtxn = index.read_txn().unwrap();
+
+            let configs = index.embedding_configs(&rtxn).unwrap();
+            // for consistency with the below
+            #[allow(clippy::get_first)]
+            let (name, fakerest_config) = configs.get(0).unwrap();
+            insta::assert_json_snapshot!(name, @r###""A_fakerest""###);
+            insta::assert_json_snapshot!(fakerest_config.embedder_options);
+            let fakerest_name = name.clone();
+
+            let (name, simple_hf_config) = configs.get(1).unwrap();
+            insta::assert_json_snapshot!(name, @r###""B_small_hf""###);
+            insta::assert_json_snapshot!(simple_hf_config.embedder_options);
+            let simple_hf_name = name.clone();
+
+            let configs = index_scheduler.embedders(configs).unwrap();
+            let (hf_embedder, _) = configs.get(&simple_hf_name).unwrap();
+            let beagle_embed = hf_embedder.embed_one(S("Intel the beagle best doggo")).unwrap();
+            let lab_embed = hf_embedder.embed_one(S("Max the lab best doggo")).unwrap();
+            let patou_embed = hf_embedder.embed_one(S("kefir the patou best doggo")).unwrap();
+            (fakerest_name, simple_hf_name, beagle_embed, lab_embed, patou_embed)
+        };
+
+        // add one doc, specifying vectors
+
+        let doc = serde_json::json!(
+                    {
+                        "id": 0,
+                        "doggo": "Intel",
+                        "breed": "beagle",
+                        "_vectors": {
+                            &fakerest_name: {
+                                // this will never trigger regeneration, which is good because we can't actually generate with
+                                // this embedder
+                                "userProvided": true,
+                                "embeddings": beagle_embed,
+                            },
+                            &simple_hf_name: {
+                                // this will be regenerated on updates
+                                "userProvided": false,
+                                "embeddings": lab_embed,
+                            },
+                            "noise": [0.1, 0.2, 0.3]
+                        }
+                    }
+        );
+
+        let (uuid, mut file) = index_scheduler.create_update_file_with_uuid(0u128).unwrap();
+        let documents_count = read_json(doc.to_string().as_bytes(), &mut file).unwrap();
+        assert_eq!(documents_count, 1);
+        file.persist().unwrap();
+
+        index_scheduler
+            .register(
+                KindWithContent::DocumentAdditionOrUpdate {
+                    index_uid: S("doggos"),
+                    primary_key: Some(S("id")),
+                    method: UpdateDocuments,
+                    content_file: uuid,
+                    documents_count,
+                    allow_index_creation: true,
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        index_scheduler.assert_internally_consistent();
+
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after adding Intel");
+
+        handle.advance_one_successful_batch();
+
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "adding Intel succeeds");
+
+        // check embeddings
+        {
+            let index = index_scheduler.index("doggos").unwrap();
+            let rtxn = index.read_txn().unwrap();
+
+            let embeddings = index.embeddings(&rtxn, 0).unwrap();
+
+            assert_json_snapshot!(embeddings[&simple_hf_name][0] == lab_embed, @"true");
+            assert_json_snapshot!(embeddings[&fakerest_name][0] == beagle_embed, @"true");
+
+            let doc = index.documents(&rtxn, std::iter::once(0)).unwrap()[0].1;
+            let fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
+            let doc = obkv_to_json(
+                &[
+                    fields_ids_map.id("doggo").unwrap(),
+                    fields_ids_map.id("breed").unwrap(),
+                    fields_ids_map.id("_vectors").unwrap(),
+                ],
+                &fields_ids_map,
+                doc,
+            )
+            .unwrap();
+            assert_json_snapshot!(doc, {"._vectors.A_fakerest.embeddings" => "[vector]"});
+        }
+
+        // update the doc, specifying vectors
+
+        let doc = serde_json::json!(
+                    {
+                        "id": 0,
+                        "doggo": "kefir",
+                        "breed": "patou",
+                    }
+        );
+
+        let (uuid, mut file) = index_scheduler.create_update_file_with_uuid(1u128).unwrap();
+        let documents_count = read_json(doc.to_string().as_bytes(), &mut file).unwrap();
+        assert_eq!(documents_count, 1);
+        file.persist().unwrap();
+
+        index_scheduler
+            .register(
+                KindWithContent::DocumentAdditionOrUpdate {
+                    index_uid: S("doggos"),
+                    primary_key: None,
+                    method: UpdateDocuments,
+                    content_file: uuid,
+                    documents_count,
+                    allow_index_creation: true,
+                },
+                None,
+                false,
+            )
+            .unwrap();
+        index_scheduler.assert_internally_consistent();
+
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "Intel to kefir");
+
+        handle.advance_one_successful_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "Intel to kefir succeeds");
+
+        {
+            // check embeddings
+            {
+                let index = index_scheduler.index("doggos").unwrap();
+                let rtxn = index.read_txn().unwrap();
+
+                let embeddings = index.embeddings(&rtxn, 0).unwrap();
+
+                // automatically changed to patou
+                assert_json_snapshot!(embeddings[&simple_hf_name][0] == patou_embed, @"true");
+                // remained beagle because set to userProvided
+                assert_json_snapshot!(embeddings[&fakerest_name][0] == beagle_embed, @"true");
+
+                let doc = index.documents(&rtxn, std::iter::once(0)).unwrap()[0].1;
+                let fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
+                let doc = obkv_to_json(
+                    &[
+                        fields_ids_map.id("doggo").unwrap(),
+                        fields_ids_map.id("breed").unwrap(),
+                        fields_ids_map.id("_vectors").unwrap(),
+                    ],
+                    &fields_ids_map,
+                    doc,
+                )
+                .unwrap();
+                assert_json_snapshot!(doc, {"._vectors.A_fakerest.embeddings" => "[vector]"});
+            }
+        }
+    }
 }
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-12.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-12.snap
@ -0,0 +1,19 @@
+---
+source: index-scheduler/src/lib.rs
+expression: doc
+---
+{
+  "doggo": "kefir",
+  "breed": "patou",
+  "_vectors": {
+    "A_fakerest": {
+      "embeddings": "[vector]",
+      "userProvided": true
+    },
+    "noise": [
+      0.1,
+      0.2,
+      0.3
+    ]
+  }
+}
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-2.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-2.snap
@ -0,0 +1,20 @@
+---
+source: index-scheduler/src/lib.rs
+expression: task.details
+---
+{
+  "embedders": {
+    "A_fakerest": {
+      "source": "rest",
+      "apiKey": "MyXXXX...",
+      "dimensions": 384,
+      "url": "http://localhost:7777"
+    },
+    "B_small_hf": {
+      "source": "huggingFace",
+      "model": "sentence-transformers/all-MiniLM-L6-v2",
+      "revision": "e4ce9877abf3edfe10b0d82785e83bdcb973e22e",
+      "documentTemplate": "{{doc.doggo}} the {{doc.breed}} best doggo"
+    }
+  }
+}
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-4.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-4.snap
@ -0,0 +1,23 @@
+---
+source: index-scheduler/src/lib.rs
+expression: fakerest_config.embedder_options
+---
+{
+  "Rest": {
+    "api_key": "My super secret",
+    "distribution": null,
+    "dimensions": 384,
+    "url": "http://localhost:7777",
+    "query": null,
+    "input_field": [
+      "input"
+    ],
+    "path_to_embeddings": [
+      "data"
+    ],
+    "embedding_object": [
+      "embedding"
+    ],
+    "input_type": "text"
+  }
+}
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-6.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-6.snap
@ -0,0 +1,11 @@
+---
+source: index-scheduler/src/lib.rs
+expression: simple_hf_config.embedder_options
+---
+{
+  "HuggingFace": {
+    "model": "sentence-transformers/all-MiniLM-L6-v2",
+    "revision": "e4ce9877abf3edfe10b0d82785e83bdcb973e22e",
+    "distribution": null
+  }
+}
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-9.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors-9.snap
@ -0,0 +1,19 @@
+---
+source: index-scheduler/src/lib.rs
+expression: doc
+---
+{
+  "doggo": "Intel",
+  "breed": "beagle",
+  "_vectors": {
+    "A_fakerest": {
+      "embeddings": "[vector]",
+      "userProvided": true
+    },
+    "noise": [
+      0.1,
+      0.2,
+      0.3
+    ]
+  }
+}
--- a/index-scheduler/src/snapshots/index_schedulertestsimport_vectors.snap
+++ b/index-scheduler/src/snapshots/index_schedulertestsimport_vectors.snap
@ -0,0 +1,20 @@
+---
+source: index-scheduler/src/lib.rs
+expression: task.details
+---
+{
+  "embedders": {
+    "A_fakerest": {
+      "source": "rest",
+      "apiKey": "MyXXXX...",
+      "dimensions": 384,
+      "url": "http://localhost:7777"
+    },
+    "B_small_hf": {
+      "source": "huggingFace",
+      "model": "sentence-transformers/all-MiniLM-L6-v2",
+      "revision": "e4ce9877abf3edfe10b0d82785e83bdcb973e22e",
+      "documentTemplate": "{{doc.doggo}} the {{doc.breed}} best doggo"
+    }
+  }
+}
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/Intel
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/Intel
@ -0,0 +1,49 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
+2 {uid: 2, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: None, method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,1,2,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,2,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,2,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1, "doggo": 1, "id": 1} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/Intel
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/Intel
@ -0,0 +1,48 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
+2 {uid: 2, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: None, method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [2,]
+succeeded [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,2,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,2,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1, "doggo": 1, "id": 1} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+[timestamp] [2,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000001
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/adding
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/adding
@ -0,0 +1,45 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 1, field_distribution: {"_vectors": 1, "breed": 1, "doggo": 1, "id": 1} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/after
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/after
@ -0,0 +1,44 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+1 {uid: 1, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [1,]
+succeeded [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 0, field_distribution: {} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/after_registering_settings_task_vectors.snap
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/after_registering_settings_task_vectors.snap
@ -0,0 +1,36 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,]
+----------------------------------------------------------------------
+### Kind:
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/snapshots/lib.rs/import_vectors/settings_update_processed_vectors.snap
+++ b/index-scheduler/src/snapshots/lib.rs/import_vectors/settings_update_processed_vectors.snap
@ -0,0 +1,40 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: NotSet, sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: Set({"A_fakerest": Set(EmbeddingSettings { source: Set(Rest), model: NotSet, revision: NotSet, api_key: Set("My super secret"), dimensions: Set(384), document_template: NotSet, url: Set("http://localhost:7777"), query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet }), "B_small_hf": Set(EmbeddingSettings { source: Set(HuggingFace), model: Set("sentence-transformers/all-MiniLM-L6-v2"), revision: Set("e4ce9877abf3edfe10b0d82785e83bdcb973e22e"), api_key: NotSet, dimensions: NotSet, document_template: Set("{{doc.doggo}} the {{doc.breed}} best doggo"), url: NotSet, query: NotSet, input_field: NotSet, path_to_embeddings: NotSet, embedding_object: NotSet, input_type: NotSet, distribution: NotSet })}), search_cutoff_ms: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,]
+----------------------------------------------------------------------
+### Kind:
+"settingsUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 0, field_distribution: {} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+
--- a/index-scheduler/src/utils.rs
+++ b/index-scheduler/src/utils.rs
@ -272,9 +272,9 @@ pub fn swap_index_uid_in_task(task: &mut Task, swap: (&str, &str)) {
    }
    for index_uid in index_uids {
        if index_uid == swap.0 {
-            *index_uid = swap.1.to_owned();
+            swap.1.clone_into(index_uid);
        } else if index_uid == swap.1 {
-            *index_uid = swap.0.to_owned();
+            swap.0.clone_into(index_uid);
        }
    }
 }
--- a/meilisearch-auth/src/store.rs
+++ b/meilisearch-auth/src/store.rs
@ -49,7 +49,7 @@ pub fn open_auth_store_env(path: &Path) -> milli::heed::Result<milli::heed::Env>
    let mut options = EnvOpenOptions::new();
    options.map_size(AUTH_STORE_SIZE); // 1GB
    options.max_dbs(2);
-    options.open(path)
+    unsafe { options.open(path) }
 }

 impl HeedAuthStore {
--- a/meilisearch-types/src/deserr/mod.rs
+++ b/meilisearch-types/src/deserr/mod.rs
@ -189,3 +189,4 @@ merge_with_error_impl_take_error_message!(ParseTaskKindError);
 merge_with_error_impl_take_error_message!(ParseTaskStatusError);
 merge_with_error_impl_take_error_message!(IndexUidFormatError);
 merge_with_error_impl_take_error_message!(InvalidSearchSemanticRatio);
+merge_with_error_impl_take_error_message!(InvalidSimilarId);
--- a/meilisearch-types/src/error.rs
+++ b/meilisearch-types/src/error.rs
@ -239,18 +239,23 @@ InvalidIndexUid                       , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchAttributesToSearchOn     , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchAttributesToCrop         , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchAttributesToHighlight    , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarAttributesToRetrieve    , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchAttributesToRetrieve     , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchCropLength               , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchCropMarker               , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFacets                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchSemanticRatio            , InvalidRequest       , BAD_REQUEST ;
 InvalidFacetSearchFacetName           , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarId                      , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchFilter                   , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarFilter                  , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHighlightPostTag         , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHighlightPreTag          , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchHitsPerPage              , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarLimit                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchLimit                    , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchMatchingStrategy         , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarOffset                  , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchOffset                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchPage                     , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchQ                        , InvalidRequest       , BAD_REQUEST ;
@ -259,7 +264,9 @@ InvalidFacetSearchName                , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchVector                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchShowMatchesPosition      , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchShowRankingScore         , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarShowRankingScore        , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchShowRankingScoreDetails  , InvalidRequest       , BAD_REQUEST ;
+InvalidSimilarShowRankingScoreDetails , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchSort                     , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsDisplayedAttributes    , InvalidRequest       , BAD_REQUEST ;
 InvalidSettingsDistinctAttribute      , InvalidRequest       , BAD_REQUEST ;
@ -322,7 +329,8 @@ UnretrievableErrorCode                , InvalidRequest       , BAD_REQUEST ;
 UnsupportedMediaType                  , InvalidRequest       , UNSUPPORTED_MEDIA_TYPE ;

 // Experimental features
-VectorEmbeddingError                  , InvalidRequest       , BAD_REQUEST
+VectorEmbeddingError                  , InvalidRequest       , BAD_REQUEST ;
+NotFoundSimilarId                     , InvalidRequest       , BAD_REQUEST
 }

 impl ErrorCode for JoinError {
@ -384,7 +392,6 @@ impl ErrorCode for milli::Error {
                    UserError::InvalidGeoField { .. } => Code::InvalidDocumentGeoField,
                    UserError::InvalidVectorDimensions { .. } => Code::InvalidVectorDimensions,
                    UserError::InvalidVectorsMapType { .. } => Code::InvalidVectorsType,
-                    UserError::InvalidVectorsType { .. } => Code::InvalidVectorsType,
                    UserError::TooManyVectors(_, _) => Code::TooManyVectors,
                    UserError::SortError(_) => Code::InvalidSearchSort,
                    UserError::InvalidMinTypoWordLenSetting(_, _) => {
@ -423,7 +430,6 @@ impl ErrorCode for HeedError {
            HeedError::Mdb(_)
            | HeedError::Encoding(_)
            | HeedError::Decoding(_)
-            | HeedError::InvalidDatabaseTyping
            | HeedError::DatabaseClosing
            | HeedError::BadOpenOptions { .. } => Code::Internal,
        }
@ -488,6 +494,17 @@ impl fmt::Display for deserr_codes::InvalidSearchSemanticRatio {
    }
 }

+impl fmt::Display for deserr_codes::InvalidSimilarId {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(
+            f,
+            "the value of `id` is invalid. \
+            A document identifier can be of type integer or string, \
+            only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_)."
+        )
+    }
+}
+
 #[macro_export]
 macro_rules! internal_error {
    ($target:ty : $($other:path), *) => {
--- a/meilisearch-types/src/features.rs
+++ b/meilisearch-types/src/features.rs
@ -6,7 +6,6 @@ pub struct RuntimeTogglableFeatures {
    pub vector_store: bool,
    pub metrics: bool,
    pub logs_route: bool,
-    pub export_puffin_reports: bool,
 }

 #[derive(Default, Debug, Clone, Copy)]
--- a/meilisearch/Cargo.toml
+++ b/meilisearch/Cargo.toml
@ -67,7 +67,6 @@ permissive-json-pointer = { path = "../permissive-json-pointer" }
 pin-project-lite = "0.2.13"
 platform-dirs = "0.3.0"
 prometheus = { version = "0.13.3", features = ["process"] }
-puffin = { version = "0.16.0", features = ["serialization"] }
 rand = "0.8.5"
 rayon = "1.8.0"
 regex = "1.10.2"
--- a/meilisearch/src/analytics/mock_analytics.rs
+++ b/meilisearch/src/analytics/mock_analytics.rs
@ -25,6 +25,18 @@ impl SearchAggregator {
    pub fn succeed(&mut self, _: &dyn Any) {}
 }

+#[derive(Default)]
+pub struct SimilarAggregator;
+
+#[allow(dead_code)]
+impl SimilarAggregator {
+    pub fn from_query(_: &dyn Any, _: &dyn Any) -> Self {
+        Self
+    }
+
+    pub fn succeed(&mut self, _: &dyn Any) {}
+}
+
 #[derive(Default)]
 pub struct MultiSearchAggregator;

@ -66,6 +78,8 @@ impl Analytics for MockAnalytics {
    fn publish(&self, _event_name: String, _send: Value, _request: Option<&HttpRequest>) {}
    fn get_search(&self, _aggregate: super::SearchAggregator) {}
    fn post_search(&self, _aggregate: super::SearchAggregator) {}
+    fn get_similar(&self, _aggregate: super::SimilarAggregator) {}
+    fn post_similar(&self, _aggregate: super::SimilarAggregator) {}
    fn post_multi_search(&self, _aggregate: super::MultiSearchAggregator) {}
    fn post_facet_search(&self, _aggregate: super::FacetSearchAggregator) {}
    fn add_documents(
--- a/meilisearch/src/analytics/mod.rs
+++ b/meilisearch/src/analytics/mod.rs
@ -22,6 +22,8 @@ pub type SegmentAnalytics = mock_analytics::MockAnalytics;
 #[cfg(not(feature = "analytics"))]
 pub type SearchAggregator = mock_analytics::SearchAggregator;
 #[cfg(not(feature = "analytics"))]
+pub type SimilarAggregator = mock_analytics::SimilarAggregator;
+#[cfg(not(feature = "analytics"))]
 pub type MultiSearchAggregator = mock_analytics::MultiSearchAggregator;
 #[cfg(not(feature = "analytics"))]
 pub type FacetSearchAggregator = mock_analytics::FacetSearchAggregator;
@ -32,6 +34,8 @@ pub type SegmentAnalytics = segment_analytics::SegmentAnalytics;
 #[cfg(feature = "analytics")]
 pub type SearchAggregator = segment_analytics::SearchAggregator;
 #[cfg(feature = "analytics")]
+pub type SimilarAggregator = segment_analytics::SimilarAggregator;
+#[cfg(feature = "analytics")]
 pub type MultiSearchAggregator = segment_analytics::MultiSearchAggregator;
 #[cfg(feature = "analytics")]
 pub type FacetSearchAggregator = segment_analytics::FacetSearchAggregator;
@ -86,6 +90,12 @@ pub trait Analytics: Sync + Send {
    /// This method should be called to aggregate a post search
    fn post_search(&self, aggregate: SearchAggregator);

+    /// This method should be called to aggregate a get similar request
+    fn get_similar(&self, aggregate: SimilarAggregator);
+
+    /// This method should be called to aggregate a post similar request
+    fn post_similar(&self, aggregate: SimilarAggregator);
+
    /// This method should be called to aggregate a post array of searches
    fn post_multi_search(&self, aggregate: MultiSearchAggregator);

--- a/meilisearch/src/analytics/segment_analytics.rs
+++ b/meilisearch/src/analytics/segment_analytics.rs
@ -36,8 +36,9 @@ use crate::routes::indexes::facet_search::FacetSearchQuery;
 use crate::routes::{create_all_stats, Stats};
 use crate::search::{
    FacetSearchResult, MatchingStrategy, SearchQuery, SearchQueryWithIndex, SearchResult,
-    DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER, DEFAULT_HIGHLIGHT_POST_TAG,
-    DEFAULT_HIGHLIGHT_PRE_TAG, DEFAULT_SEARCH_LIMIT, DEFAULT_SEMANTIC_RATIO,
+    SimilarQuery, SimilarResult, DEFAULT_CROP_LENGTH, DEFAULT_CROP_MARKER,
+    DEFAULT_HIGHLIGHT_POST_TAG, DEFAULT_HIGHLIGHT_PRE_TAG, DEFAULT_SEARCH_LIMIT,
+    DEFAULT_SEMANTIC_RATIO,
 };
 use crate::Opt;

@ -73,6 +74,8 @@ pub enum AnalyticsMsg {
    BatchMessage(Track),
    AggregateGetSearch(SearchAggregator),
    AggregatePostSearch(SearchAggregator),
+    AggregateGetSimilar(SimilarAggregator),
+    AggregatePostSimilar(SimilarAggregator),
    AggregatePostMultiSearch(MultiSearchAggregator),
    AggregatePostFacetSearch(FacetSearchAggregator),
    AggregateAddDocuments(DocumentsAggregator),
@ -149,6 +152,8 @@ impl SegmentAnalytics {
            update_documents_aggregator: DocumentsAggregator::default(),
            get_fetch_documents_aggregator: DocumentsFetchAggregator::default(),
            post_fetch_documents_aggregator: DocumentsFetchAggregator::default(),
+            get_similar_aggregator: SimilarAggregator::default(),
+            post_similar_aggregator: SimilarAggregator::default(),
        });
        tokio::spawn(segment.run(index_scheduler.clone(), auth_controller.clone()));

@ -184,6 +189,14 @@ impl super::Analytics for SegmentAnalytics {
        let _ = self.sender.try_send(AnalyticsMsg::AggregatePostSearch(aggregate));
    }

+    fn get_similar(&self, aggregate: SimilarAggregator) {
+        let _ = self.sender.try_send(AnalyticsMsg::AggregateGetSimilar(aggregate));
+    }
+
+    fn post_similar(&self, aggregate: SimilarAggregator) {
+        let _ = self.sender.try_send(AnalyticsMsg::AggregatePostSimilar(aggregate));
+    }
+
    fn post_facet_search(&self, aggregate: FacetSearchAggregator) {
        let _ = self.sender.try_send(AnalyticsMsg::AggregatePostFacetSearch(aggregate));
    }
@ -379,6 +392,8 @@ pub struct Segment {
    update_documents_aggregator: DocumentsAggregator,
    get_fetch_documents_aggregator: DocumentsFetchAggregator,
    post_fetch_documents_aggregator: DocumentsFetchAggregator,
+    get_similar_aggregator: SimilarAggregator,
+    post_similar_aggregator: SimilarAggregator,
 }

 impl Segment {
@ -441,6 +456,8 @@ impl Segment {
                        Some(AnalyticsMsg::AggregateUpdateDocuments(agreg)) => self.update_documents_aggregator.aggregate(agreg),
                        Some(AnalyticsMsg::AggregateGetFetchDocuments(agreg)) => self.get_fetch_documents_aggregator.aggregate(agreg),
                        Some(AnalyticsMsg::AggregatePostFetchDocuments(agreg)) => self.post_fetch_documents_aggregator.aggregate(agreg),
+                        Some(AnalyticsMsg::AggregateGetSimilar(agreg)) => self.get_similar_aggregator.aggregate(agreg),
+                        Some(AnalyticsMsg::AggregatePostSimilar(agreg)) => self.post_similar_aggregator.aggregate(agreg),
                        None => (),
                    }
                }
@ -494,6 +511,8 @@ impl Segment {
            update_documents_aggregator,
            get_fetch_documents_aggregator,
            post_fetch_documents_aggregator,
+            get_similar_aggregator,
+            post_similar_aggregator,
        } = self;

        if let Some(get_search) =
@ -541,6 +560,18 @@ impl Segment {
        {
            let _ = self.batcher.push(post_fetch_documents).await;
        }
+
+        if let Some(get_similar_documents) =
+            take(get_similar_aggregator).into_event(user, "Similar GET")
+        {
+            let _ = self.batcher.push(get_similar_documents).await;
+        }
+
+        if let Some(post_similar_documents) =
+            take(post_similar_aggregator).into_event(user, "Similar POST")
+        {
+            let _ = self.batcher.push(post_similar_documents).await;
+        }
        let _ = self.batcher.flush().await;
    }
 }
@ -1558,3 +1589,235 @@ impl DocumentsFetchAggregator {
        })
    }
 }
+
+#[derive(Default)]
+pub struct SimilarAggregator {
+    timestamp: Option<OffsetDateTime>,
+
+    // context
+    user_agents: HashSet<String>,
+
+    // requests
+    total_received: usize,
+    total_succeeded: usize,
+    time_spent: BinaryHeap<usize>,
+
+    // filter
+    filter_with_geo_radius: bool,
+    filter_with_geo_bounding_box: bool,
+    // every time a request has a filter, this field must be incremented by the number of terms it contains
+    filter_sum_of_criteria_terms: usize,
+    // every time a request has a filter, this field must be incremented by one
+    filter_total_number_of_criteria: usize,
+    used_syntax: HashMap<String, usize>,
+
+    // Whether a non-default embedder was specified
+    embedder: bool,
+
+    // pagination
+    max_limit: usize,
+    max_offset: usize,
+
+    // formatting
+    max_attributes_to_retrieve: usize,
+
+    // scoring
+    show_ranking_score: bool,
+    show_ranking_score_details: bool,
+}
+
+impl SimilarAggregator {
+    #[allow(clippy::field_reassign_with_default)]
+    pub fn from_query(query: &SimilarQuery, request: &HttpRequest) -> Self {
+        let SimilarQuery {
+            id: _,
+            embedder,
+            offset,
+            limit,
+            attributes_to_retrieve: _,
+            show_ranking_score,
+            show_ranking_score_details,
+            filter,
+        } = query;
+
+        let mut ret = Self::default();
+        ret.timestamp = Some(OffsetDateTime::now_utc());
+
+        ret.total_received = 1;
+        ret.user_agents = extract_user_agents(request).into_iter().collect();
+
+        if let Some(ref filter) = filter {
+            static RE: Lazy<Regex> = Lazy::new(|| Regex::new("AND | OR").unwrap());
+            ret.filter_total_number_of_criteria = 1;
+
+            let syntax = match filter {
+                Value::String(_) => "string".to_string(),
+                Value::Array(values) => {
+                    if values.iter().map(|v| v.to_string()).any(|s| RE.is_match(&s)) {
+                        "mixed".to_string()
+                    } else {
+                        "array".to_string()
+                    }
+                }
+                _ => "none".to_string(),
+            };
+            // convert the string to a HashMap
+            ret.used_syntax.insert(syntax, 1);
+
+            let stringified_filters = filter.to_string();
+            ret.filter_with_geo_radius = stringified_filters.contains("_geoRadius(");
+            ret.filter_with_geo_bounding_box = stringified_filters.contains("_geoBoundingBox(");
+            ret.filter_sum_of_criteria_terms = RE.split(&stringified_filters).count();
+        }
+
+        ret.max_limit = *limit;
+        ret.max_offset = *offset;
+
+        ret.show_ranking_score = *show_ranking_score;
+        ret.show_ranking_score_details = *show_ranking_score_details;
+
+        ret.embedder = embedder.is_some();
+
+        ret
+    }
+
+    pub fn succeed(&mut self, result: &SimilarResult) {
+        let SimilarResult { id: _, hits: _, processing_time_ms, hits_info: _ } = result;
+
+        self.total_succeeded = self.total_succeeded.saturating_add(1);
+
+        self.time_spent.push(*processing_time_ms as usize);
+    }
+
+    /// Aggregate one [SimilarAggregator] into another.
+    pub fn aggregate(&mut self, mut other: Self) {
+        let Self {
+            timestamp,
+            user_agents,
+            total_received,
+            total_succeeded,
+            ref mut time_spent,
+            filter_with_geo_radius,
+            filter_with_geo_bounding_box,
+            filter_sum_of_criteria_terms,
+            filter_total_number_of_criteria,
+            used_syntax,
+            max_limit,
+            max_offset,
+            max_attributes_to_retrieve,
+            show_ranking_score,
+            show_ranking_score_details,
+            embedder,
+        } = other;
+
+        if self.timestamp.is_none() {
+            self.timestamp = timestamp;
+        }
+
+        // context
+        for user_agent in user_agents.into_iter() {
+            self.user_agents.insert(user_agent);
+        }
+
+        // request
+        self.total_received = self.total_received.saturating_add(total_received);
+        self.total_succeeded = self.total_succeeded.saturating_add(total_succeeded);
+        self.time_spent.append(time_spent);
+
+        // filter
+        self.filter_with_geo_radius |= filter_with_geo_radius;
+        self.filter_with_geo_bounding_box |= filter_with_geo_bounding_box;
+        self.filter_sum_of_criteria_terms =
+            self.filter_sum_of_criteria_terms.saturating_add(filter_sum_of_criteria_terms);
+        self.filter_total_number_of_criteria =
+            self.filter_total_number_of_criteria.saturating_add(filter_total_number_of_criteria);
+        for (key, value) in used_syntax.into_iter() {
+            let used_syntax = self.used_syntax.entry(key).or_insert(0);
+            *used_syntax = used_syntax.saturating_add(value);
+        }
+
+        self.embedder |= embedder;
+
+        // pagination
+        self.max_limit = self.max_limit.max(max_limit);
+        self.max_offset = self.max_offset.max(max_offset);
+
+        // formatting
+        self.max_attributes_to_retrieve =
+            self.max_attributes_to_retrieve.max(max_attributes_to_retrieve);
+
+        // scoring
+        self.show_ranking_score |= show_ranking_score;
+        self.show_ranking_score_details |= show_ranking_score_details;
+    }
+
+    pub fn into_event(self, user: &User, event_name: &str) -> Option<Track> {
+        let Self {
+            timestamp,
+            user_agents,
+            total_received,
+            total_succeeded,
+            time_spent,
+            filter_with_geo_radius,
+            filter_with_geo_bounding_box,
+            filter_sum_of_criteria_terms,
+            filter_total_number_of_criteria,
+            used_syntax,
+            max_limit,
+            max_offset,
+            max_attributes_to_retrieve,
+            show_ranking_score,
+            show_ranking_score_details,
+            embedder,
+        } = self;
+
+        if total_received == 0 {
+            None
+        } else {
+            // we get all the values in a sorted manner
+            let time_spent = time_spent.into_sorted_vec();
+            // the index of the 99th percentage of value
+            let percentile_99th = time_spent.len() * 99 / 100;
+            // We are only interested by the slowest value of the 99th fastest results
+            let time_spent = time_spent.get(percentile_99th);
+
+            let properties = json!({
+                "user-agent": user_agents,
+                "requests": {
+                    "99th_response_time": time_spent.map(|t| format!("{:.2}", t)),
+                    "total_succeeded": total_succeeded,
+                    "total_failed": total_received.saturating_sub(total_succeeded), // just to be sure we never panics
+                    "total_received": total_received,
+                },
+                "filter": {
+                   "with_geoRadius": filter_with_geo_radius,
+                   "with_geoBoundingBox": filter_with_geo_bounding_box,
+                   "avg_criteria_number": format!("{:.2}", filter_sum_of_criteria_terms as f64 / filter_total_number_of_criteria as f64),
+                   "most_used_syntax": used_syntax.iter().max_by_key(|(_, v)| *v).map(|(k, _)| json!(k)).unwrap_or_else(|| json!(null)),
+                },
+                "hybrid": {
+                    "embedder": embedder,
+                },
+                "pagination": {
+                   "max_limit": max_limit,
+                   "max_offset": max_offset,
+                },
+                "formatting": {
+                    "max_attributes_to_retrieve": max_attributes_to_retrieve,
+                },
+                "scoring": {
+                    "show_ranking_score": show_ranking_score,
+                    "show_ranking_score_details": show_ranking_score_details,
+                },
+            });
+
+            Some(Track {
+                timestamp,
+                user: user.clone(),
+                event: event_name.to_string(),
+                properties,
+                ..Default::default()
+            })
+        }
+    }
+}
--- a/meilisearch/src/routes/features.rs
+++ b/meilisearch/src/routes/features.rs
@ -47,8 +47,6 @@ pub struct RuntimeTogglableFeatures {
    pub metrics: Option<bool>,
    #[deserr(default)]
    pub logs_route: Option<bool>,
-    #[deserr(default)]
-    pub export_puffin_reports: Option<bool>,
 }

 async fn patch_features(
@ -68,21 +66,13 @@ async fn patch_features(
        vector_store: new_features.0.vector_store.unwrap_or(old_features.vector_store),
        metrics: new_features.0.metrics.unwrap_or(old_features.metrics),
        logs_route: new_features.0.logs_route.unwrap_or(old_features.logs_route),
-        export_puffin_reports: new_features
-            .0
-            .export_puffin_reports
-            .unwrap_or(old_features.export_puffin_reports),
    };

    // explicitly destructure for analytics rather than using the `Serialize` implementation, because
    // the it renames to camelCase, which we don't want for analytics.
    // **Do not** ignore fields with `..` or `_` here, because we want to add them in the future.
-    let meilisearch_types::features::RuntimeTogglableFeatures {
-        vector_store,
-        metrics,
-        logs_route,
-        export_puffin_reports,
-    } = new_features;
+    let meilisearch_types::features::RuntimeTogglableFeatures { vector_store, metrics, logs_route } =
+        new_features;

    analytics.publish(
        "Experimental features Updated".to_string(),
@ -90,7 +80,6 @@ async fn patch_features(
            "vector_store": vector_store,
            "metrics": metrics,
            "logs_route": logs_route,
-            "export_puffin_reports": export_puffin_reports,
        }),
        Some(&req),
    );
--- a/meilisearch/src/routes/indexes/facet_search.rs
+++ b/meilisearch/src/routes/indexes/facet_search.rs
@ -69,7 +69,7 @@ pub async fn search(

    // Tenant token search_rules.
    if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
-        add_search_rules(&mut search_query, search_rules);
+        add_search_rules(&mut search_query.filter, search_rules);
    }

    let index = index_scheduler.index(&index_uid)?;
--- a/meilisearch/src/routes/indexes/mod.rs
+++ b/meilisearch/src/routes/indexes/mod.rs
@ -29,6 +29,7 @@ pub mod documents;
 pub mod facet_search;
 pub mod search;
 pub mod settings;
+pub mod similar;

 pub fn configure(cfg: &mut web::ServiceConfig) {
    cfg.service(
@ -45,9 +46,14 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
                    .route(web::delete().to(SeqHandler(delete_index))),
            )
            .service(web::resource("/stats").route(web::get().to(SeqHandler(get_index_stats))))
+            .service(
+                web::resource("/advanced-stats")
+                    .route(web::get().to(SeqHandler(get_advanced_index_stats))),
+            )
            .service(web::scope("/documents").configure(documents::configure))
            .service(web::scope("/search").configure(search::configure))
            .service(web::scope("/facet-search").configure(facet_search::configure))
+            .service(web::scope("/similar").configure(similar::configure))
            .service(web::scope("/settings").configure(settings::configure)),
    );
 }
@ -276,3 +282,16 @@ pub async fn get_index_stats(
    debug!(returns = ?stats, "Get index stats");
    Ok(HttpResponse::Ok().json(stats))
 }
+
+pub async fn get_advanced_index_stats(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::STATS_GET }>, Data<IndexScheduler>>,
+    index_uid: web::Path<String>,
+) -> Result<HttpResponse, ResponseError> {
+    let index_uid = IndexUid::try_from(index_uid.into_inner())?;
+    let index = index_scheduler.index(&index_uid)?;
+    let rtxn = index.read_txn()?;
+    let advanced_stats = index.advanced_stats(&rtxn)?;
+
+    debug!(returns = ?advanced_stats, "Get advanced index stats");
+    Ok(HttpResponse::Ok().json(advanced_stats))
+}
--- a/meilisearch/src/routes/indexes/search.rs
+++ b/meilisearch/src/routes/indexes/search.rs
@ -196,7 +196,7 @@ pub async fn search_with_url_query(

    // Tenant token search_rules.
    if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
-        add_search_rules(&mut query, search_rules);
+        add_search_rules(&mut query.filter, search_rules);
    }

    let mut aggregate = SearchAggregator::from_query(&query, &req);
@ -235,7 +235,7 @@ pub async fn search_with_post(

    // Tenant token search_rules.
    if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
-        add_search_rules(&mut query, search_rules);
+        add_search_rules(&mut query.filter, search_rules);
    }

    let mut aggregate = SearchAggregator::from_query(&query, &req);
--- a/meilisearch/src/routes/indexes/similar.rs
+++ b/meilisearch/src/routes/indexes/similar.rs
@ -0,0 +1,171 @@
+use actix_web::web::{self, Data};
+use actix_web::{HttpRequest, HttpResponse};
+use deserr::actix_web::{AwebJson, AwebQueryParameter};
+use index_scheduler::IndexScheduler;
+use meilisearch_types::deserr::query_params::Param;
+use meilisearch_types::deserr::{DeserrJsonError, DeserrQueryParamError};
+use meilisearch_types::error::deserr_codes::{
+    InvalidEmbedder, InvalidSimilarAttributesToRetrieve, InvalidSimilarFilter, InvalidSimilarId,
+    InvalidSimilarLimit, InvalidSimilarOffset, InvalidSimilarShowRankingScore,
+    InvalidSimilarShowRankingScoreDetails,
+};
+use meilisearch_types::error::{ErrorCode as _, ResponseError};
+use meilisearch_types::index_uid::IndexUid;
+use meilisearch_types::keys::actions;
+use meilisearch_types::serde_cs::vec::CS;
+use serde_json::Value;
+use tracing::debug;
+
+use super::ActionPolicy;
+use crate::analytics::{Analytics, SimilarAggregator};
+use crate::extractors::authentication::GuardedData;
+use crate::extractors::sequential_extractor::SeqHandler;
+use crate::search::{
+    add_search_rules, perform_similar, SearchKind, SimilarQuery, SimilarResult,
+    DEFAULT_SEARCH_LIMIT, DEFAULT_SEARCH_OFFSET,
+};
+
+pub fn configure(cfg: &mut web::ServiceConfig) {
+    cfg.service(
+        web::resource("")
+            .route(web::get().to(SeqHandler(similar_get)))
+            .route(web::post().to(SeqHandler(similar_post))),
+    );
+}
+
+pub async fn similar_get(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
+    index_uid: web::Path<String>,
+    params: AwebQueryParameter<SimilarQueryGet, DeserrQueryParamError>,
+    req: HttpRequest,
+    analytics: web::Data<dyn Analytics>,
+) -> Result<HttpResponse, ResponseError> {
+    let index_uid = IndexUid::try_from(index_uid.into_inner())?;
+
+    let query = params.0.try_into().map_err(|code: InvalidSimilarId| {
+        ResponseError::from_msg(code.to_string(), code.error_code())
+    })?;
+
+    let mut aggregate = SimilarAggregator::from_query(&query, &req);
+
+    debug!(parameters = ?query, "Similar get");
+
+    let similar = similar(index_scheduler, index_uid, query).await;
+
+    if let Ok(similar) = &similar {
+        aggregate.succeed(similar);
+    }
+    analytics.get_similar(aggregate);
+
+    let similar = similar?;
+
+    debug!(returns = ?similar, "Similar get");
+    Ok(HttpResponse::Ok().json(similar))
+}
+
+pub async fn similar_post(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
+    index_uid: web::Path<String>,
+    params: AwebJson<SimilarQuery, DeserrJsonError>,
+    req: HttpRequest,
+    analytics: web::Data<dyn Analytics>,
+) -> Result<HttpResponse, ResponseError> {
+    let index_uid = IndexUid::try_from(index_uid.into_inner())?;
+
+    let query = params.into_inner();
+    debug!(parameters = ?query, "Similar post");
+
+    let mut aggregate = SimilarAggregator::from_query(&query, &req);
+
+    let similar = similar(index_scheduler, index_uid, query).await;
+
+    if let Ok(similar) = &similar {
+        aggregate.succeed(similar);
+    }
+    analytics.post_similar(aggregate);
+
+    let similar = similar?;
+
+    debug!(returns = ?similar, "Similar post");
+    Ok(HttpResponse::Ok().json(similar))
+}
+
+async fn similar(
+    index_scheduler: GuardedData<ActionPolicy<{ actions::SEARCH }>, Data<IndexScheduler>>,
+    index_uid: IndexUid,
+    mut query: SimilarQuery,
+) -> Result<SimilarResult, ResponseError> {
+    let features = index_scheduler.features();
+
+    features.check_vector("Using the similar API")?;
+
+    // Tenant token search_rules.
+    if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid) {
+        add_search_rules(&mut query.filter, search_rules);
+    }
+
+    let index = index_scheduler.index(&index_uid)?;
+
+    let (embedder_name, embedder) =
+        SearchKind::embedder(&index_scheduler, &index, query.embedder.as_deref(), None)?;
+
+    tokio::task::spawn_blocking(move || perform_similar(&index, query, embedder_name, embedder))
+        .await?
+}
+
+#[derive(Debug, deserr::Deserr)]
+#[deserr(error = DeserrQueryParamError, rename_all = camelCase, deny_unknown_fields)]
+pub struct SimilarQueryGet {
+    #[deserr(error = DeserrQueryParamError<InvalidSimilarId>)]
+    id: Param<String>,
+    #[deserr(default = Param(DEFAULT_SEARCH_OFFSET()), error = DeserrQueryParamError<InvalidSimilarOffset>)]
+    offset: Param<usize>,
+    #[deserr(default = Param(DEFAULT_SEARCH_LIMIT()), error = DeserrQueryParamError<InvalidSimilarLimit>)]
+    limit: Param<usize>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSimilarAttributesToRetrieve>)]
+    attributes_to_retrieve: Option<CS<String>>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSimilarFilter>)]
+    filter: Option<String>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSimilarShowRankingScore>)]
+    show_ranking_score: Param<bool>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidSimilarShowRankingScoreDetails>)]
+    show_ranking_score_details: Param<bool>,
+    #[deserr(default, error = DeserrQueryParamError<InvalidEmbedder>)]
+    pub embedder: Option<String>,
+}
+
+impl TryFrom<SimilarQueryGet> for SimilarQuery {
+    type Error = InvalidSimilarId;
+
+    fn try_from(
+        SimilarQueryGet {
+            id,
+            offset,
+            limit,
+            attributes_to_retrieve,
+            filter,
+            show_ranking_score,
+            show_ranking_score_details,
+            embedder,
+        }: SimilarQueryGet,
+    ) -> Result<Self, Self::Error> {
+        let filter = match filter {
+            Some(f) => match serde_json::from_str(&f) {
+                Ok(v) => Some(v),
+                _ => Some(Value::String(f)),
+            },
+            None => None,
+        };
+
+        Ok(SimilarQuery {
+            id: id.0.try_into()?,
+            offset: offset.0,
+            limit: limit.0,
+            filter,
+            embedder,
+            attributes_to_retrieve: attributes_to_retrieve.map(|o| o.into_iter().collect()),
+            show_ranking_score: show_ranking_score.0,
+            show_ranking_score_details: show_ranking_score_details.0,
+        })
+    }
+}
--- a/meilisearch/src/routes/multi_search.rs
+++ b/meilisearch/src/routes/multi_search.rs
@ -67,7 +67,7 @@ pub async fn multi_search_with_post(
            // Apply search rules from tenant token
            if let Some(search_rules) = index_scheduler.filters().get_index_search_rules(&index_uid)
            {
-                add_search_rules(&mut query, search_rules);
+                add_search_rules(&mut query.filter, search_rules);
            }

            let index = index_scheduler
--- a/meilisearch/src/search.rs
+++ b/meilisearch/src/search.rs
@ -11,7 +11,7 @@ use indexmap::IndexMap;
 use meilisearch_auth::IndexSearchRules;
 use meilisearch_types::deserr::DeserrJsonError;
 use meilisearch_types::error::deserr_codes::*;
-use meilisearch_types::error::ResponseError;
+use meilisearch_types::error::{Code, ResponseError};
 use meilisearch_types::heed::RoTxn;
 use meilisearch_types::index_uid::IndexUid;
 use meilisearch_types::milli::score_details::{ScoreDetails, ScoringStrategy};
@ -231,7 +231,7 @@ impl SearchKind {
        Ok(Self::Hybrid { embedder_name, embedder, semantic_ratio })
    }

-    fn embedder(
+    pub(crate) fn embedder(
        index_scheduler: &index_scheduler::IndexScheduler,
        index: &Index,
        embedder_name: Option<&str>,
@ -417,6 +417,59 @@ impl SearchQueryWithIndex {
    }
 }

+#[derive(Debug, Clone, PartialEq, Deserr)]
+#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
+pub struct SimilarQuery {
+    #[deserr(error = DeserrJsonError<InvalidSimilarId>)]
+    pub id: ExternalDocumentId,
+    #[deserr(default = DEFAULT_SEARCH_OFFSET(), error = DeserrJsonError<InvalidSimilarOffset>)]
+    pub offset: usize,
+    #[deserr(default = DEFAULT_SEARCH_LIMIT(), error = DeserrJsonError<InvalidSimilarLimit>)]
+    pub limit: usize,
+    #[deserr(default, error = DeserrJsonError<InvalidSimilarFilter>)]
+    pub filter: Option<Value>,
+    #[deserr(default, error = DeserrJsonError<InvalidEmbedder>, default)]
+    pub embedder: Option<String>,
+    #[deserr(default, error = DeserrJsonError<InvalidSimilarAttributesToRetrieve>)]
+    pub attributes_to_retrieve: Option<BTreeSet<String>>,
+    #[deserr(default, error = DeserrJsonError<InvalidSimilarShowRankingScore>, default)]
+    pub show_ranking_score: bool,
+    #[deserr(default, error = DeserrJsonError<InvalidSimilarShowRankingScoreDetails>, default)]
+    pub show_ranking_score_details: bool,
+}
+
+#[derive(Debug, Clone, PartialEq, Deserr)]
+#[deserr(try_from(Value) = TryFrom::try_from -> InvalidSimilarId)]
+pub struct ExternalDocumentId(String);
+
+impl AsRef<str> for ExternalDocumentId {
+    fn as_ref(&self) -> &str {
+        &self.0
+    }
+}
+
+impl ExternalDocumentId {
+    pub fn into_inner(self) -> String {
+        self.0
+    }
+}
+
+impl TryFrom<String> for ExternalDocumentId {
+    type Error = InvalidSimilarId;
+
+    fn try_from(value: String) -> Result<Self, Self::Error> {
+        serde_json::Value::String(value).try_into()
+    }
+}
+
+impl TryFrom<Value> for ExternalDocumentId {
+    type Error = InvalidSimilarId;
+
+    fn try_from(value: Value) -> Result<Self, Self::Error> {
+        Ok(Self(milli::documents::validate_document_id_value(value).map_err(|_| InvalidSimilarId)?))
+    }
+}
+
 #[derive(Debug, Copy, Clone, PartialEq, Eq, Deserr)]
 #[deserr(rename_all = camelCase)]
 pub enum MatchingStrategy {
@ -424,6 +477,8 @@ pub enum MatchingStrategy {
    Last,
    /// All query words are mandatory
    All,
+    /// Remove query words from the most frequent to the least
+    Frequency,
 }

 impl Default for MatchingStrategy {
@ -437,6 +492,7 @@ impl From<MatchingStrategy> for TermsMatchingStrategy {
        match other {
            MatchingStrategy::Last => Self::Last,
            MatchingStrategy::All => Self::All,
+            MatchingStrategy::Frequency => Self::Frequency,
        }
    }
 }
@ -538,6 +594,16 @@ impl fmt::Debug for SearchResult {
    }
 }

+#[derive(Serialize, Debug, Clone, PartialEq)]
+#[serde(rename_all = "camelCase")]
+pub struct SimilarResult {
+    pub hits: Vec<SearchHit>,
+    pub id: String,
+    pub processing_time_ms: u128,
+    #[serde(flatten)]
+    pub hits_info: HitsInfo,
+}
+
 #[derive(Serialize, Debug, Clone, PartialEq)]
 #[serde(rename_all = "camelCase")]
 pub struct SearchResultWithIndex {
@ -570,8 +636,8 @@ pub struct FacetSearchResult {
 }

 /// Incorporate search rules in search query
-pub fn add_search_rules(query: &mut SearchQuery, rules: IndexSearchRules) {
-    query.filter = match (query.filter.take(), rules.filter) {
+pub fn add_search_rules(filter: &mut Option<Value>, rules: IndexSearchRules) {
+    *filter = match (filter.take(), rules.filter) {
        (None, rules_filter) => rules_filter,
        (filter, None) => filter,
        (Some(filter), Some(rules_filter)) => {
@ -719,131 +785,52 @@ pub fn perform_search(
        SearchKind::Hybrid { semantic_ratio, .. } => search.execute_hybrid(*semantic_ratio)?,
    };

-    let fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
+    let SearchQuery {
+        q,
+        vector: _,
+        hybrid: _,
+        // already computed from prepare_search
+        offset: _,
+        limit,
+        page,
+        hits_per_page,
+        attributes_to_retrieve,
+        attributes_to_crop,
+        crop_length,
+        attributes_to_highlight,
+        show_matches_position,
+        show_ranking_score,
+        show_ranking_score_details,
+        filter: _,
+        sort,
+        facets,
+        highlight_pre_tag,
+        highlight_post_tag,
+        crop_marker,
+        matching_strategy: _,
+        attributes_to_search_on: _,
+    } = query;

-    let displayed_ids = index
-        .displayed_fields_ids(&rtxn)?
-        .map(|fields| fields.into_iter().collect::<BTreeSet<_>>())
-        .unwrap_or_else(|| fields_ids_map.iter().map(|(id, _)| id).collect());
-
-    let fids = |attrs: &BTreeSet<String>| {
-        let mut ids = BTreeSet::new();
-        for attr in attrs {
-            if attr == "*" {
-                ids = displayed_ids.clone();
-                break;
-            }
-
-            if let Some(id) = fields_ids_map.id(attr) {
-                ids.insert(id);
-            }
-        }
-        ids
+    let format = AttributesFormat {
+        attributes_to_retrieve,
+        attributes_to_highlight,
+        attributes_to_crop,
+        crop_length,
+        crop_marker,
+        highlight_pre_tag,
+        highlight_post_tag,
+        show_matches_position,
+        sort,
+        show_ranking_score,
+        show_ranking_score_details,
    };

-    // The attributes to retrieve are the ones explicitly marked as to retrieve (all by default),
-    // but these attributes must be also be present
-    // - in the fields_ids_map
-    // - in the displayed attributes
-    let to_retrieve_ids: BTreeSet<_> = query
-        .attributes_to_retrieve
-        .as_ref()
-        .map(fids)
-        .unwrap_or_else(|| displayed_ids.clone())
-        .intersection(&displayed_ids)
-        .cloned()
-        .collect();
-
-    let attr_to_highlight = query.attributes_to_highlight.unwrap_or_default();
-
-    let attr_to_crop = query.attributes_to_crop.unwrap_or_default();
-
-    // Attributes in `formatted_options` correspond to the attributes that will be in `_formatted`
-    // These attributes are:
-    // - the attributes asked to be highlighted or cropped (with `attributesToCrop` or `attributesToHighlight`)
-    // - the attributes asked to be retrieved: these attributes will not be highlighted/cropped
-    // But these attributes must be also present in displayed attributes
-    let formatted_options = compute_formatted_options(
-        &attr_to_highlight,
-        &attr_to_crop,
-        query.crop_length,
-        &to_retrieve_ids,
-        &fields_ids_map,
-        &displayed_ids,
-    );
-
-    let mut tokenizer_builder = TokenizerBuilder::default();
-    tokenizer_builder.create_char_map(true);
-
-    let script_lang_map = index.script_language(&rtxn)?;
-    if !script_lang_map.is_empty() {
-        tokenizer_builder.allow_list(&script_lang_map);
-    }
-
-    let separators = index.allowed_separators(&rtxn)?;
-    let separators: Option<Vec<_>> =
-        separators.as_ref().map(|x| x.iter().map(String::as_str).collect());
-    if let Some(ref separators) = separators {
-        tokenizer_builder.separators(separators);
-    }
-
-    let dictionary = index.dictionary(&rtxn)?;
-    let dictionary: Option<Vec<_>> =
-        dictionary.as_ref().map(|x| x.iter().map(String::as_str).collect());
-    if let Some(ref dictionary) = dictionary {
-        tokenizer_builder.words_dict(dictionary);
-    }
-
-    let mut formatter_builder = MatcherBuilder::new(matching_words, tokenizer_builder.build());
-    formatter_builder.crop_marker(query.crop_marker);
-    formatter_builder.highlight_prefix(query.highlight_pre_tag);
-    formatter_builder.highlight_suffix(query.highlight_post_tag);
-
-    let mut documents = Vec::new();
-    let documents_iter = index.documents(&rtxn, documents_ids)?;
-
-    for ((_id, obkv), score) in documents_iter.into_iter().zip(document_scores.into_iter()) {
-        // First generate a document with all the displayed fields
-        let displayed_document = make_document(&displayed_ids, &fields_ids_map, obkv)?;
-
-        // select the attributes to retrieve
-        let attributes_to_retrieve = to_retrieve_ids
-            .iter()
-            .map(|&fid| fields_ids_map.name(fid).expect("Missing field name"));
-        let mut document =
-            permissive_json_pointer::select_values(&displayed_document, attributes_to_retrieve);
-
-        let (matches_position, formatted) = format_fields(
-            &displayed_document,
-            &fields_ids_map,
-            &formatter_builder,
-            &formatted_options,
-            query.show_matches_position,
-            &displayed_ids,
-        )?;
-
-        if let Some(sort) = query.sort.as_ref() {
-            insert_geo_distance(sort, &mut document);
-        }
-
-        let ranking_score =
-            query.show_ranking_score.then(|| ScoreDetails::global_score(score.iter()));
-        let ranking_score_details =
-            query.show_ranking_score_details.then(|| ScoreDetails::to_json_map(score.iter()));
-
-        let hit = SearchHit {
-            document,
-            formatted,
-            matches_position,
-            ranking_score_details,
-            ranking_score,
-        };
-        documents.push(hit);
-    }
+    let documents =
+        make_hits(index, &rtxn, format, matching_words, documents_ids, document_scores)?;

    let number_of_hits = min(candidates.len() as usize, max_total_hits);
    let hits_info = if is_finite_pagination {
-        let hits_per_page = query.hits_per_page.unwrap_or_else(DEFAULT_SEARCH_LIMIT);
+        let hits_per_page = hits_per_page.unwrap_or_else(DEFAULT_SEARCH_LIMIT);
        // If hit_per_page is 0, then pages can't be computed and so we respond 0.
        let total_pages = (number_of_hits + hits_per_page.saturating_sub(1))
            .checked_div(hits_per_page)
@ -851,15 +838,15 @@ pub fn perform_search(

        HitsInfo::Pagination {
            hits_per_page,
-            page: query.page.unwrap_or(1),
+            page: page.unwrap_or(1),
            total_pages,
            total_hits: number_of_hits,
        }
    } else {
-        HitsInfo::OffsetLimit { limit: query.limit, offset, estimated_total_hits: number_of_hits }
+        HitsInfo::OffsetLimit { limit, offset, estimated_total_hits: number_of_hits }
    };

-    let (facet_distribution, facet_stats) = match query.facets {
+    let (facet_distribution, facet_stats) = match facets {
        Some(ref fields) => {
            let mut facet_distribution = index.facets_distribution(&rtxn);

@ -896,7 +883,7 @@ pub fn perform_search(
    let result = SearchResult {
        hits: documents,
        hits_info,
-        query: query.q.unwrap_or_default(),
+        query: q.unwrap_or_default(),
        processing_time_ms: before_search.elapsed().as_millis(),
        facet_distribution,
        facet_stats,
@ -907,6 +894,130 @@ pub fn perform_search(
    Ok(result)
 }

+struct AttributesFormat {
+    attributes_to_retrieve: Option<BTreeSet<String>>,
+    attributes_to_highlight: Option<HashSet<String>>,
+    attributes_to_crop: Option<Vec<String>>,
+    crop_length: usize,
+    crop_marker: String,
+    highlight_pre_tag: String,
+    highlight_post_tag: String,
+    show_matches_position: bool,
+    sort: Option<Vec<String>>,
+    show_ranking_score: bool,
+    show_ranking_score_details: bool,
+}
+
+fn make_hits(
+    index: &Index,
+    rtxn: &RoTxn<'_>,
+    format: AttributesFormat,
+    matching_words: milli::MatchingWords,
+    documents_ids: Vec<u32>,
+    document_scores: Vec<Vec<ScoreDetails>>,
+) -> Result<Vec<SearchHit>, MeilisearchHttpError> {
+    let fields_ids_map = index.fields_ids_map(rtxn).unwrap();
+    let displayed_ids = index
+        .displayed_fields_ids(rtxn)?
+        .map(|fields| fields.into_iter().collect::<BTreeSet<_>>())
+        .unwrap_or_else(|| fields_ids_map.iter().map(|(id, _)| id).collect());
+    let fids = |attrs: &BTreeSet<String>| {
+        let mut ids = BTreeSet::new();
+        for attr in attrs {
+            if attr == "*" {
+                ids.clone_from(&displayed_ids);
+                break;
+            }
+
+            if let Some(id) = fields_ids_map.id(attr) {
+                ids.insert(id);
+            }
+        }
+        ids
+    };
+    let to_retrieve_ids: BTreeSet<_> = format
+        .attributes_to_retrieve
+        .as_ref()
+        .map(fids)
+        .unwrap_or_else(|| displayed_ids.clone())
+        .intersection(&displayed_ids)
+        .cloned()
+        .collect();
+    let attr_to_highlight = format.attributes_to_highlight.unwrap_or_default();
+    let attr_to_crop = format.attributes_to_crop.unwrap_or_default();
+    let formatted_options = compute_formatted_options(
+        &attr_to_highlight,
+        &attr_to_crop,
+        format.crop_length,
+        &to_retrieve_ids,
+        &fields_ids_map,
+        &displayed_ids,
+    );
+    let mut tokenizer_builder = TokenizerBuilder::default();
+    tokenizer_builder.create_char_map(true);
+    let script_lang_map = index.script_language(rtxn)?;
+    if !script_lang_map.is_empty() {
+        tokenizer_builder.allow_list(&script_lang_map);
+    }
+    let separators = index.allowed_separators(rtxn)?;
+    let separators: Option<Vec<_>> =
+        separators.as_ref().map(|x| x.iter().map(String::as_str).collect());
+    if let Some(ref separators) = separators {
+        tokenizer_builder.separators(separators);
+    }
+    let dictionary = index.dictionary(rtxn)?;
+    let dictionary: Option<Vec<_>> =
+        dictionary.as_ref().map(|x| x.iter().map(String::as_str).collect());
+    if let Some(ref dictionary) = dictionary {
+        tokenizer_builder.words_dict(dictionary);
+    }
+    let mut formatter_builder = MatcherBuilder::new(matching_words, tokenizer_builder.build());
+    formatter_builder.crop_marker(format.crop_marker);
+    formatter_builder.highlight_prefix(format.highlight_pre_tag);
+    formatter_builder.highlight_suffix(format.highlight_post_tag);
+    let mut documents = Vec::new();
+    let documents_iter = index.documents(rtxn, documents_ids)?;
+    for ((_id, obkv), score) in documents_iter.into_iter().zip(document_scores.into_iter()) {
+        // First generate a document with all the displayed fields
+        let displayed_document = make_document(&displayed_ids, &fields_ids_map, obkv)?;
+
+        // select the attributes to retrieve
+        let attributes_to_retrieve = to_retrieve_ids
+            .iter()
+            .map(|&fid| fields_ids_map.name(fid).expect("Missing field name"));
+        let mut document =
+            permissive_json_pointer::select_values(&displayed_document, attributes_to_retrieve);
+
+        let (matches_position, formatted) = format_fields(
+            &displayed_document,
+            &fields_ids_map,
+            &formatter_builder,
+            &formatted_options,
+            format.show_matches_position,
+            &displayed_ids,
+        )?;
+
+        if let Some(sort) = format.sort.as_ref() {
+            insert_geo_distance(sort, &mut document);
+        }
+
+        let ranking_score =
+            format.show_ranking_score.then(|| ScoreDetails::global_score(score.iter()));
+        let ranking_score_details =
+            format.show_ranking_score_details.then(|| ScoreDetails::to_json_map(score.iter()));
+
+        let hit = SearchHit {
+            document,
+            formatted,
+            matches_position,
+            ranking_score_details,
+            ranking_score,
+        };
+        documents.push(hit);
+    }
+    Ok(documents)
+}
+
 pub fn perform_facet_search(
    index: &Index,
    search_query: SearchQuery,
@ -941,6 +1052,95 @@ pub fn perform_facet_search(
    })
 }

+pub fn perform_similar(
+    index: &Index,
+    query: SimilarQuery,
+    embedder_name: String,
+    embedder: Arc<Embedder>,
+) -> Result<SimilarResult, ResponseError> {
+    let before_search = Instant::now();
+    let rtxn = index.read_txn()?;
+
+    let SimilarQuery {
+        id,
+        offset,
+        limit,
+        filter: _,
+        embedder: _,
+        attributes_to_retrieve,
+        show_ranking_score,
+        show_ranking_score_details,
+    } = query;
+
+    // using let-else rather than `?` so that the borrow checker identifies we're always returning here,
+    // preventing a use-after-move
+    let Some(internal_id) = index.external_documents_ids().get(&rtxn, &id)? else {
+        return Err(ResponseError::from_msg(
+            MeilisearchHttpError::DocumentNotFound(id.into_inner()).to_string(),
+            Code::NotFoundSimilarId,
+        ));
+    };
+
+    let mut similar =
+        milli::Similar::new(internal_id, offset, limit, index, &rtxn, embedder_name, embedder);
+
+    if let Some(ref filter) = query.filter {
+        if let Some(facets) = parse_filter(filter)
+            // inject InvalidSimilarFilter code
+            .map_err(|e| ResponseError::from_msg(e.to_string(), Code::InvalidSimilarFilter))?
+        {
+            similar.filter(facets);
+        }
+    }
+
+    let milli::SearchResult {
+        documents_ids,
+        matching_words: _,
+        candidates,
+        document_scores,
+        degraded: _,
+        used_negative_operator: _,
+    } = similar.execute().map_err(|err| match err {
+        milli::Error::UserError(milli::UserError::InvalidFilter(_)) => {
+            ResponseError::from_msg(err.to_string(), Code::InvalidSimilarFilter)
+        }
+        err => err.into(),
+    })?;
+
+    let format = AttributesFormat {
+        attributes_to_retrieve,
+        attributes_to_highlight: None,
+        attributes_to_crop: None,
+        crop_length: DEFAULT_CROP_LENGTH(),
+        crop_marker: DEFAULT_CROP_MARKER(),
+        highlight_pre_tag: DEFAULT_HIGHLIGHT_PRE_TAG(),
+        highlight_post_tag: DEFAULT_HIGHLIGHT_POST_TAG(),
+        show_matches_position: false,
+        sort: None,
+        show_ranking_score,
+        show_ranking_score_details,
+    };
+
+    let hits = make_hits(index, &rtxn, format, Default::default(), documents_ids, document_scores)?;
+
+    let max_total_hits = index
+        .pagination_max_total_hits(&rtxn)
+        .map_err(milli::Error::from)?
+        .map(|x| x as usize)
+        .unwrap_or(DEFAULT_PAGINATION_MAX_TOTAL_HITS);
+
+    let number_of_hits = min(candidates.len() as usize, max_total_hits);
+    let hits_info = HitsInfo::OffsetLimit { limit, offset, estimated_total_hits: number_of_hits };
+
+    let result = SimilarResult {
+        hits,
+        hits_info,
+        id: id.into_inner(),
+        processing_time_ms: before_search.elapsed().as_millis(),
+    };
+    Ok(result)
+}
+
 fn insert_geo_distance(sorts: &[String], document: &mut Document) {
    lazy_static::lazy_static! {
        static ref GEO_REGEX: Regex =
--- a/meilisearch/src/search_queue.rs
+++ b/meilisearch/src/search_queue.rs
@ -85,8 +85,13 @@ impl SearchQueue {
                },

                search_request = receive_new_searches.recv() => {
-                    // this unwrap is safe because we're sure the `SearchQueue` still lives somewhere in actix-web
-                    let search_request = search_request.unwrap();
+                    let search_request = match search_request {
+                        Some(search_request) => search_request,
+                        // This should never happen while actix-web is running, but it's not a reason to crash
+                        // and it can generate a lot of noise in the tests.
+                        None => continue,
+                    };
+
                    if searches_running < usize::from(parallelism) && queue.is_empty() {
                        searches_running += 1;
                        // if the search requests die it's not a hard error on our side
--- a/meilisearch/tests/common/index.rs
+++ b/meilisearch/tests/common/index.rs
@ -380,6 +380,43 @@ impl Index<'_> {
        self.service.get(url).await
    }

+    /// Performs both GET and POST similar queries
+    pub async fn similar(
+        &self,
+        query: Value,
+        test: impl Fn(Value, StatusCode) + UnwindSafe + Clone,
+    ) {
+        let post = self.similar_post(query.clone()).await;
+
+        let query = yaup::to_string(&query).unwrap();
+        let get = self.similar_get(&query).await;
+
+        insta::allow_duplicates! {
+            let (response, code) = post;
+            let t = test.clone();
+            if let Err(e) = catch_unwind(move || t(response, code)) {
+                eprintln!("Error with post search");
+                resume_unwind(e);
+            }
+
+            let (response, code) = get;
+            if let Err(e) = catch_unwind(move || test(response, code)) {
+                eprintln!("Error with get search");
+                resume_unwind(e);
+            }
+        }
+    }
+
+    pub async fn similar_post(&self, query: Value) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/similar", urlencode(self.uid.as_ref()));
+        self.service.post_encoded(url, query, self.encoder).await
+    }
+
+    pub async fn similar_get(&self, query: &str) -> (Value, StatusCode) {
+        let url = format!("/indexes/{}/similar?{}", urlencode(self.uid.as_ref()), query);
+        self.service.get(url).await
+    }
+
    pub async fn facet_search(&self, query: Value) -> (Value, StatusCode) {
        let url = format!("/indexes/{}/facet-search", urlencode(self.uid.as_ref()));
        self.service.post_encoded(url, query, self.encoder).await
--- a/meilisearch/tests/dumps/mod.rs
+++ b/meilisearch/tests/dumps/mod.rs
@ -1859,8 +1859,7 @@ async fn import_dump_v6_containing_experimental_features() {
    {
      "vectorStore": false,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

--- a/meilisearch/tests/features/mod.rs
+++ b/meilisearch/tests/features/mod.rs
@ -20,8 +20,7 @@ async fn experimental_features() {
    {
      "vectorStore": false,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -32,8 +31,7 @@ async fn experimental_features() {
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -44,8 +42,7 @@ async fn experimental_features() {
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -57,8 +54,7 @@ async fn experimental_features() {
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -70,8 +66,7 @@ async fn experimental_features() {
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);
 }
@ -90,8 +85,7 @@ async fn experimental_feature_metrics() {
    {
      "vectorStore": false,
      "metrics": true,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -146,7 +140,7 @@ async fn errors() {
    meili_snap::snapshot!(code, @"400 Bad Request");
    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
    {
-      "message": "Unknown field `NotAFeature`: expected one of `vectorStore`, `metrics`, `logsRoute`, `exportPuffinReports`",
+      "message": "Unknown field `NotAFeature`: expected one of `vectorStore`, `metrics`, `logsRoute`",
      "code": "bad_request",
      "type": "invalid_request",
      "link": "https://docs.meilisearch.com/errors#bad_request"
--- a/meilisearch/tests/integration.rs
+++ b/meilisearch/tests/integration.rs
@ -8,6 +8,7 @@ mod index;
 mod logs;
 mod search;
 mod settings;
+mod similar;
 mod snapshot;
 mod stats;
 mod swap_indexes;
--- a/meilisearch/tests/search/errors.rs
+++ b/meilisearch/tests/search/errors.rs
@ -505,7 +505,7 @@ async fn search_bad_matching_strategy() {
    snapshot!(code, @"400 Bad Request");
    snapshot!(json_string!(response), @r###"
    {
-      "message": "Unknown value `doggo` at `.matchingStrategy`: expected one of `last`, `all`",
+      "message": "Unknown value `doggo` at `.matchingStrategy`: expected one of `last`, `all`, `frequency`",
      "code": "invalid_search_matching_strategy",
      "type": "invalid_request",
      "link": "https://docs.meilisearch.com/errors#invalid_search_matching_strategy"
@ -527,7 +527,7 @@ async fn search_bad_matching_strategy() {
    snapshot!(code, @"400 Bad Request");
    snapshot!(json_string!(response), @r###"
    {
-      "message": "Unknown value `doggo` for parameter `matchingStrategy`: expected one of `last`, `all`",
+      "message": "Unknown value `doggo` for parameter `matchingStrategy`: expected one of `last`, `all`, `frequency`",
      "code": "invalid_search_matching_strategy",
      "type": "invalid_request",
      "link": "https://docs.meilisearch.com/errors#invalid_search_matching_strategy"
--- a/meilisearch/tests/search/geo.rs
+++ b/meilisearch/tests/search/geo.rs
@ -117,3 +117,69 @@ async fn geo_bounding_box_with_string_and_number() {
        )
        .await;
 }
+
+#[actix_rt::test]
+async fn bug_4640() {
+    // https://github.com/meilisearch/meilisearch/issues/4640
+    let server = Server::new().await;
+    let index = server.index("test");
+
+    let documents = DOCUMENTS.clone();
+    index.add_documents(documents, None).await;
+    index.update_settings_filterable_attributes(json!(["_geo"])).await;
+    let (ret, _code) = index.update_settings_sortable_attributes(json!(["_geo"])).await;
+    index.wait_task(ret.uid()).await;
+
+    // Sort the document with the second one first
+    index
+        .search(
+            json!({
+                "sort": ["_geoPoint(45.4777599, 9.1967508):asc"],
+            }),
+            |response, code| {
+                assert_eq!(code, 200, "{}", response);
+                snapshot!(json_string!(response, { ".processingTimeMs" => "[time]" }), @r###"
+                {
+                  "hits": [
+                    {
+                      "id": 2,
+                      "name": "La Bella Italia",
+                      "address": "456 Elm Street, Townsville",
+                      "type": "Italian",
+                      "rating": 9,
+                      "_geo": {
+                        "lat": "45.4777599",
+                        "lng": "9.1967508"
+                      }
+                    },
+                    {
+                      "id": 1,
+                      "name": "Taco Truck",
+                      "address": "444 Salsa Street, Burritoville",
+                      "type": "Mexican",
+                      "rating": 9,
+                      "_geo": {
+                        "lat": 34.0522,
+                        "lng": -118.2437
+                      },
+                      "_geoDistance": 9714063
+                    },
+                    {
+                      "id": 3,
+                      "name": "Crêpe Truck",
+                      "address": "2 Billig Avenue, Rouenville",
+                      "type": "French",
+                      "rating": 10
+                    }
+                  ],
+                  "query": "",
+                  "processingTimeMs": "[time]",
+                  "limit": 20,
+                  "offset": 0,
+                  "estimatedTotalHits": 3
+                }
+                "###);
+            },
+        )
+        .await;
+}
--- a/meilisearch/tests/search/hybrid.rs
+++ b/meilisearch/tests/search/hybrid.rs
@ -5,7 +5,10 @@ use crate::common::index::Index;
 use crate::common::{Server, Value};
 use crate::json;

-async fn index_with_documents<'a>(server: &'a Server, documents: &Value) -> Index<'a> {
+async fn index_with_documents_user_provided<'a>(
+    server: &'a Server,
+    documents: &Value,
+) -> Index<'a> {
    let index = server.index("test");

    let (response, code) = server.set_features(json!({"vectorStore": true})).await;
@ -15,8 +18,7 @@ async fn index_with_documents<'a>(server: &'a Server, documents: &Value) -> Inde
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

@ -34,7 +36,38 @@ async fn index_with_documents<'a>(server: &'a Server, documents: &Value) -> Inde
    index
 }

-static SIMPLE_SEARCH_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+async fn index_with_documents_hf<'a>(server: &'a Server, documents: &Value) -> Index<'a> {
+    let index = server.index("test");
+
+    let (response, code) = server.set_features(json!({"vectorStore": true})).await;
+
+    meili_snap::snapshot!(code, @"200 OK");
+    meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
+    {
+      "vectorStore": true,
+      "metrics": false,
+      "logsRoute": false
+    }
+    "###);
+
+    let (response, code) = index
+        .update_settings(json!({ "embedders": {"default": {
+            "source": "huggingFace",
+            "model": "sentence-transformers/all-MiniLM-L6-v2",
+            "revision": "e4ce9877abf3edfe10b0d82785e83bdcb973e22e",
+            "documentTemplate": "{{doc.title}}, {{doc.desc}}"
+        }}} ))
+        .await;
+    assert_eq!(202, code, "{:?}", response);
+    index.wait_task(response.uid()).await;
+
+    let (response, code) = index.add_documents(documents.clone(), None).await;
+    assert_eq!(202, code, "{:?}", response);
+    index.wait_task(response.uid()).await;
+    index
+}
+
+static SIMPLE_SEARCH_DOCUMENTS_VEC: Lazy<Value> = Lazy::new(|| {
    json!([
    {
        "title": "Shazam!",
@ -56,7 +89,7 @@ static SIMPLE_SEARCH_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
    }])
 });

-static SINGLE_DOCUMENT: Lazy<Value> = Lazy::new(|| {
+static SINGLE_DOCUMENT_VEC: Lazy<Value> = Lazy::new(|| {
    json!([{
            "title": "Shazam!",
            "desc": "a Captain Marvel ersatz",
@ -65,10 +98,29 @@ static SINGLE_DOCUMENT: Lazy<Value> = Lazy::new(|| {
    }])
 });

+static SIMPLE_SEARCH_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+    {
+        "title": "Shazam!",
+        "desc": "a Captain Marvel ersatz",
+        "id": "1",
+    },
+    {
+        "title": "Captain Planet",
+        "desc": "He's not part of the Marvel Cinematic Universe",
+        "id": "2",
+    },
+    {
+        "title": "Captain Marvel",
+        "desc": "a Shazam ersatz",
+        "id": "3",
+    }])
+});
+
 #[actix_rt::test]
 async fn simple_search() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+    let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;

    let (response, code) = index
        .search_post(
@ -85,8 +137,8 @@ async fn simple_search() {
        )
        .await;
    snapshot!(code, @"200 OK");
-    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.996969696969697},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":[2.0,3.0]},"_rankingScore":0.996969696969697},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":[1.0,3.0]},"_rankingScore":0.9472135901451112}]"###);
-    snapshot!(response["semanticHitCount"], @"1");
+    snapshot!(response["hits"], @r###"[{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":[2.0,3.0]},"_rankingScore":0.990290343761444},{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.9848484848484848},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":[1.0,3.0]},"_rankingScore":0.9472135901451112}]"###);
+    snapshot!(response["semanticHitCount"], @"2");

    let (response, code) = index
        .search_post(
@ -98,10 +150,59 @@ async fn simple_search() {
    snapshot!(response["semanticHitCount"], @"3");
 }

+#[actix_rt::test]
+async fn simple_search_hf() {
+    let server = Server::new().await;
+    let index = index_with_documents_hf(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+
+    let (response, code) =
+        index.search_post(json!({"q": "Captain", "hybrid": {"semanticRatio": 0.2}})).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"}]"###);
+    snapshot!(response["semanticHitCount"], @"0");
+
+    let (response, code) = index
+        .search_post(
+            // disable ranking score as the vectors between architectures are not equal
+            json!({"q": "Captain", "hybrid": {"semanticRatio": 0.55}, "showRankingScore": false}),
+        )
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"}]"###);
+    snapshot!(response["semanticHitCount"], @"1");
+
+    let (response, code) = index
+        .search_post(
+            json!({"q": "Captain", "hybrid": {"semanticRatio": 0.8}, "showRankingScore": false}),
+        )
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(response["hits"], @r###"[{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"}]"###);
+    snapshot!(response["semanticHitCount"], @"3");
+
+    let (response, code) = index
+        .search_post(
+            json!({"q": "Movie World", "hybrid": {"semanticRatio": 0.2}, "showRankingScore": false}),
+        )
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"}]"###);
+    snapshot!(response["semanticHitCount"], @"3");
+
+    let (response, code) = index
+        .search_post(
+            json!({"q": "Wonder replacement", "hybrid": {"semanticRatio": 0.2}, "showRankingScore": false}),
+        )
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(response["hits"], @r###"[{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3"},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1"},{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2"}]"###);
+    snapshot!(response["semanticHitCount"], @"3");
+}
+
 #[actix_rt::test]
 async fn distribution_shift() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+    let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;

    let search = json!({"q": "Captain", "vector": [1.0, 1.0], "showRankingScore": true, "hybrid": {"semanticRatio": 1.0}});
    let (response, code) = index.search_post(search.clone()).await;
@ -133,7 +234,7 @@ async fn distribution_shift() {
 #[actix_rt::test]
 async fn highlighter() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+    let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;

    let (response, code) = index
        .search_post(json!({"q": "Captain Marvel", "vector": [1.0, 1.0],
@ -184,7 +285,7 @@ async fn highlighter() {
 #[actix_rt::test]
 async fn invalid_semantic_ratio() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+    let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;

    let (response, code) = index
        .search_post(
@ -256,7 +357,7 @@ async fn invalid_semantic_ratio() {
 #[actix_rt::test]
 async fn single_document() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SINGLE_DOCUMENT).await;
+    let index = index_with_documents_user_provided(&server, &SINGLE_DOCUMENT_VEC).await;

    let (response, code) = index
    .search_post(
@ -272,7 +373,7 @@ async fn single_document() {
 #[actix_rt::test]
 async fn query_combination() {
    let server = Server::new().await;
-    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+    let index = index_with_documents_user_provided(&server, &SIMPLE_SEARCH_DOCUMENTS_VEC).await;

    // search without query and vector, but with hybrid => still placeholder
    let (response, code) = index
@ -331,7 +432,7 @@ async fn query_combination() {
    .await;

    snapshot!(code, @"200 OK");
-    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.996969696969697},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":[2.0,3.0]},"_rankingScore":0.996969696969697},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":[1.0,3.0]},"_rankingScore":0.8848484848484849}]"###);
+    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.9848484848484848},{"title":"Captain Marvel","desc":"a Shazam ersatz","id":"3","_vectors":{"default":[2.0,3.0]},"_rankingScore":0.9848484848484848},{"title":"Shazam!","desc":"a Captain Marvel ersatz","id":"1","_vectors":{"default":[1.0,3.0]},"_rankingScore":0.9242424242424242}]"###);
    snapshot!(response["semanticHitCount"], @"null");

    // query + vector, no hybrid keyword =>
@ -374,6 +475,6 @@ async fn query_combination() {
        .await;

    snapshot!(code, @"200 OK");
-    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.9848484848484848}]"###);
+    snapshot!(response["hits"], @r###"[{"title":"Captain Planet","desc":"He's not part of the Marvel Cinematic Universe","id":"2","_vectors":{"default":[1.0,2.0]},"_rankingScore":0.9242424242424242}]"###);
    snapshot!(response["semanticHitCount"], @"0");
 }
--- a/meilisearch/tests/search/matching_strategy.rs
+++ b/meilisearch/tests/search/matching_strategy.rs
@ -0,0 +1,128 @@
+use meili_snap::snapshot;
+use once_cell::sync::Lazy;
+
+use crate::common::index::Index;
+use crate::common::{Server, Value};
+use crate::json;
+
+async fn index_with_documents<'a>(server: &'a Server, documents: &Value) -> Index<'a> {
+    let index = server.index("test");
+
+    index.add_documents(documents.clone(), None).await;
+    index.wait_task(0).await;
+    index
+}
+
+static SIMPLE_SEARCH_DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+    {
+        "title": "Shazam!",
+        "id": "1",
+    },
+    {
+        "title": "Captain Planet",
+        "id": "2",
+    },
+    {
+        "title": "Captain Marvel",
+        "id": "3",
+    },
+    {
+        "title": "a Captain Marvel ersatz",
+        "id": "4"
+    },
+    {
+        "title": "He's not part of the Marvel Cinematic Universe",
+        "id": "5"
+    },
+    {
+        "title": "a Shazam ersatz, but better than Captain Planet",
+        "id": "6"
+    },
+    {
+        "title": "Capitain CAAAAAVEEERNE!!!!",
+        "id": "7"
+    }
+    ])
+});
+
+#[actix_rt::test]
+async fn simple_search() {
+    let server = Server::new().await;
+    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+
+    index
+        .search(json!({"q": "Captain Marvel", "matchingStrategy": "last", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"},{"id":"2"},{"id":"6"},{"id":"7"}]"###);
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Captain Marvel", "matchingStrategy": "all", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"}]"###);
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Captain Marvel", "matchingStrategy": "frequency", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"},{"id":"5"}]"###);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn search_with_typo() {
+    let server = Server::new().await;
+    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+
+    index
+        .search(json!({"q": "Capitain Marvel", "matchingStrategy": "last", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"},{"id":"7"},{"id":"2"},{"id":"6"}]"###);
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Capitain Marvel", "matchingStrategy": "all", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"}]"###);
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Capitain Marvel", "matchingStrategy": "frequency", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"},{"id":"5"}]"###);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn search_with_unknown_word() {
+    let server = Server::new().await;
+    let index = index_with_documents(&server, &SIMPLE_SEARCH_DOCUMENTS).await;
+
+    index
+        .search(json!({"q": "Captain Supercopter Marvel", "matchingStrategy": "last", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"2"},{"id":"3"},{"id":"4"},{"id":"6"},{"id":"7"}]"###);
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Captain Supercopter Marvel", "matchingStrategy": "all", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @"[]");
+        })
+        .await;
+
+    index
+        .search(json!({"q": "Captain Supercopter Marvel", "matchingStrategy": "frequency", "attributesToRetrieve": ["id"]}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(response["hits"], @r###"[{"id":"3"},{"id":"4"},{"id":"5"}]"###);
+        })
+        .await;
+}
--- a/meilisearch/tests/search/mod.rs
+++ b/meilisearch/tests/search/mod.rs
@ -7,6 +7,7 @@ mod facet_search;
 mod formatted;
 mod geo;
 mod hybrid;
+mod matching_strategy;
 mod multi;
 mod pagination;
 mod restrict_searchable;
@ -680,6 +681,26 @@ async fn search_facet_distribution() {
            },
        )
        .await;
+
+    index.update_settings(json!({"filterableAttributes": ["doggos.name"]})).await;
+    index.wait_task(5).await;
+
+    index
+        .search(
+            json!({
+                "facets": ["doggos.name"]
+            }),
+            |response, code| {
+                assert_eq!(code, 200, "{}", response);
+                let dist = response["facetDistribution"].as_object().unwrap();
+                assert_eq!(dist.len(), 1);
+                assert_eq!(
+                    dist["doggos.name"],
+                    json!({ "bobby": 1, "buddy": 1, "gros bill": 1, "turbo": 1, "fast": 1})
+                );
+            },
+        )
+        .await;
 }

 #[actix_rt::test]
@ -895,9 +916,9 @@ async fn test_score_details() {
                    "id": "166428",
                    "_vectors": {
                      "manual": [
-                        -100,
-                        231,
-                        32
+                        -100.0,
+                        231.0,
+                        32.0
                      ]
                    },
                    "_rankingScoreDetails": {
@ -921,7 +942,7 @@ async fn test_score_details() {
                        "order": 3,
                        "attributeRankingOrderScore": 1.0,
                        "queryWordDistanceScore": 0.8095238095238095,
-                        "score": 0.9727891156462584
+                        "score": 0.8095238095238095
                      },
                      "exactness": {
                        "order": 4,
@ -1096,9 +1117,9 @@ async fn experimental_feature_vector_store() {
        "id": "287947",
        "_vectors": {
          "manual": [
-            1,
-            2,
-            3
+            1.0,
+            2.0,
+            3.0
          ]
        },
        "_rankingScore": 1.0
@ -1108,9 +1129,9 @@ async fn experimental_feature_vector_store() {
        "id": "299537",
        "_vectors": {
          "manual": [
-            1,
-            2,
-            54
+            1.0,
+            2.0,
+            54.0
          ]
        },
        "_rankingScore": 0.9129111766815186
@ -1120,9 +1141,9 @@ async fn experimental_feature_vector_store() {
        "id": "450465",
        "_vectors": {
          "manual": [
-            -100,
-            340,
-            90
+            -100.0,
+            340.0,
+            90.0
          ]
        },
        "_rankingScore": 0.8106412887573242
@ -1132,9 +1153,9 @@ async fn experimental_feature_vector_store() {
        "id": "166428",
        "_vectors": {
          "manual": [
-            -100,
-            231,
-            32
+            -100.0,
+            231.0,
+            32.0
          ]
        },
        "_rankingScore": 0.7412010431289673
@ -1144,9 +1165,9 @@ async fn experimental_feature_vector_store() {
        "id": "522681",
        "_vectors": {
          "manual": [
-            10,
-            -23,
-            32
+            10.0,
+            -23.0,
+            32.0
          ]
        },
        "_rankingScore": 0.6972063183784485
@ -1405,9 +1426,9 @@ async fn simple_search_with_strange_synonyms() {
                "id": "166428",
                "_vectors": {
                  "manual": [
-                    -100,
-                    231,
-                    32
+                    -100.0,
+                    231.0,
+                    32.0
                  ]
                }
              }
@ -1426,9 +1447,9 @@ async fn simple_search_with_strange_synonyms() {
                "id": "166428",
                "_vectors": {
                  "manual": [
-                    -100,
-                    231,
-                    32
+                    -100.0,
+                    231.0,
+                    32.0
                  ]
                }
              }
@ -1447,9 +1468,9 @@ async fn simple_search_with_strange_synonyms() {
                "id": "166428",
                "_vectors": {
                  "manual": [
-                    -100,
-                    231,
-                    32
+                    -100.0,
+                    231.0,
+                    32.0
                  ]
                }
              }
--- a/meilisearch/tests/search/multi.rs
+++ b/meilisearch/tests/search/multi.rs
@ -75,9 +75,9 @@ async fn simple_search_single_index() {
            "id": "450465",
            "_vectors": {
              "manual": [
-                -100,
-                340,
-                90
+                -100.0,
+                340.0,
+                90.0
              ]
            }
          }
@ -96,9 +96,9 @@ async fn simple_search_single_index() {
            "id": "299537",
            "_vectors": {
              "manual": [
-                1,
-                2,
-                54
+                1.0,
+                2.0,
+                54.0
              ]
            }
          }
@ -194,9 +194,9 @@ async fn simple_search_two_indexes() {
            "id": "450465",
            "_vectors": {
              "manual": [
-                -100,
-                340,
-                90
+                -100.0,
+                340.0,
+                90.0
              ]
            }
          }
@ -227,9 +227,9 @@ async fn simple_search_two_indexes() {
            "cattos": "pésti",
            "_vectors": {
              "manual": [
-                1,
-                2,
-                3
+                1.0,
+                2.0,
+                3.0
              ]
            }
          },
@ -249,9 +249,9 @@ async fn simple_search_two_indexes() {
            ],
            "_vectors": {
              "manual": [
-                1,
-                2,
-                54
+                1.0,
+                2.0,
+                54.0
              ]
            }
          }
--- a/meilisearch/tests/search/restrict_searchable.rs
+++ b/meilisearch/tests/search/restrict_searchable.rs
@ -285,10 +285,10 @@ async fn attributes_ranking_rule_order() {
                @r###"
            [
              {
-                "id": "2"
+                "id": "1"
              },
              {
-                "id": "1"
+                "id": "2"
              }
            ]
            "###
--- a/meilisearch/tests/settings/get_settings.rs
+++ b/meilisearch/tests/settings/get_settings.rs
@ -98,8 +98,7 @@ async fn secrets_are_hidden_in_settings() {
    {
      "vectorStore": true,
      "metrics": false,
-      "logsRoute": false,
-      "exportPuffinReports": false
+      "logsRoute": false
    }
    "###);

--- a/meilisearch/tests/similar/errors.rs
+++ b/meilisearch/tests/similar/errors.rs
@ -0,0 +1,696 @@
+use meili_snap::*;
+
+use super::DOCUMENTS;
+use crate::common::Server;
+use crate::json;
+
+#[actix_rt::test]
+async fn similar_unexisting_index() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let expected_response = json!({
+        "message": "Index `test` not found.",
+        "code": "index_not_found",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#index_not_found"
+    });
+
+    index
+        .similar(json!({"id": 287947}), |response, code| {
+            assert_eq!(code, 404);
+            assert_eq!(response, expected_response);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn similar_unexisting_parameter() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    index
+        .similar(json!({"id": 287947, "marin": "hello"}), |response, code| {
+            assert_eq!(code, 400, "{}", response);
+            assert_eq!(response["code"], "bad_request");
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn similar_feature_not_enabled() {
+    let server = Server::new().await;
+    let index = server.index("test");
+
+    let (response, code) = index.similar_post(json!({"id": 287947})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Using the similar API requires enabling the `vector store` experimental feature. See https://github.com/meilisearch/product/discussions/677",
+      "code": "feature_not_enabled",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#feature_not_enabled"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_bad_id() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let (response, code) = index.similar_post(json!({"id": ["doggo"]})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_).",
+      "code": "invalid_similar_id",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_id"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_invalid_id() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let (response, code) = index.similar_post(json!({"id": "http://invalid-docid/"})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_).",
+      "code": "invalid_similar_id",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_id"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_not_found_id() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let (response, code) = index.similar_post(json!({"id": "definitely-doesnt-exist"})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Document `definitely-doesnt-exist` not found.",
+      "code": "not_found_similar_id",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#not_found_similar_id"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_bad_offset() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let (response, code) = index.similar_post(json!({"id": 287947, "offset": "doggo"})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value type at `.offset`: expected a positive integer, but found a string: `\"doggo\"`",
+      "code": "invalid_similar_offset",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_offset"
+    }
+    "###);
+
+    let (response, code) = index.similar_get("id=287947&offset=doggo").await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value in parameter `offset`: could not parse `doggo` as a positive integer",
+      "code": "invalid_similar_offset",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_offset"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_bad_limit() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let (response, code) = index.similar_post(json!({"id": 287947, "limit": "doggo"})).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value type at `.limit`: expected a positive integer, but found a string: `\"doggo\"`",
+      "code": "invalid_similar_limit",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_limit"
+    }
+    "###);
+
+    let (response, code) = index.similar_get("id=287946&limit=doggo").await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid value in parameter `limit`: could not parse `doggo` as a positive integer",
+      "code": "invalid_similar_limit",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_limit"
+    }
+    "###);
+}
+
+#[actix_rt::test]
+async fn similar_bad_filter() {
+    // Since a filter is deserialized as a json Value it will never fail to deserialize.
+    // Thus the error message is not generated by deserr but written by us.
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    snapshot!(code, @"202 Accepted");
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let (response, code) = index.similar_post(json!({ "id": 287947, "filter": true })).await;
+    snapshot!(code, @"400 Bad Request");
+    snapshot!(json_string!(response), @r###"
+    {
+      "message": "Invalid syntax for the filter parameter: `expected String, Array, found: true`.",
+      "code": "invalid_similar_filter",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    }
+    "###);
+    // Can't make the `filter` fail with a get search since it'll accept anything as a strings.
+}
+
+#[actix_rt::test]
+async fn filter_invalid_syntax_object() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": "title & Glass"}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_invalid_syntax_array() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "Was expecting an operation `=`, `!=`, `>=`, `>`, `<=`, `<`, `IN`, `NOT IN`, `TO`, `EXISTS`, `NOT EXISTS`, `IS NULL`, `IS NOT NULL`, `IS EMPTY`, `IS NOT EMPTY`, `_geoRadius`, or `_geoBoundingBox` at `title & Glass`.\n1:14 title & Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": ["title & Glass"]}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_invalid_syntax_string() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "Found unexpected characters at the end of the filter: `XOR title = Glass`. You probably forgot an `OR` or an `AND` rule.\n15:32 title = Glass XOR title = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(
+            json!({"id": 287947, "filter": "title = Glass XOR title = Glass"}),
+            |response, code| {
+                assert_eq!(response, expected_response);
+                assert_eq!(code, 400);
+            },
+        )
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_invalid_attribute_array() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": ["many = Glass"]}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_invalid_attribute_string() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": "many = Glass"}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_geo_attribute_array() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "`_geo` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:13 _geo = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": ["_geo = Glass"]}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_geo_attribute_string() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "`_geo` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:13 _geo = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": "_geo = Glass"}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_attribute_array() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "`_geoDistance` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:21 _geoDistance = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": ["_geoDistance = Glass"]}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_attribute_string() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+       "message": "`_geoDistance` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:21 _geoDistance = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": "_geoDistance = Glass"}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_geo_point_array() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+        "message": "`_geoPoint` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:18 _geoPoint = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": ["_geoPoint = Glass"]}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter_reserved_geo_point_string() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    server.set_features(json!({"vectorStore": true})).await;
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    let expected_response = json!({
+       "message": "`_geoPoint` is a reserved keyword and thus can't be used as a filter expression. Use the `_geoRadius(latitude, longitude, distance)` or `_geoBoundingBox([latitude, longitude], [latitude, longitude])` built-in rules to filter on `_geo` coordinates.\n1:18 _geoPoint = Glass",
+        "code": "invalid_similar_filter",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_similar_filter"
+    });
+    index
+        .similar(json!({"id": 287947, "filter": "_geoPoint = Glass"}), |response, code| {
+            assert_eq!(response, expected_response);
+            assert_eq!(code, 400);
+        })
+        .await;
+}
--- a/meilisearch/tests/similar/mod.rs
+++ b/meilisearch/tests/similar/mod.rs
@ -0,0 +1,373 @@
+mod errors;
+
+use meili_snap::{json_string, snapshot};
+use once_cell::sync::Lazy;
+
+use crate::common::{Server, Value};
+use crate::json;
+
+static DOCUMENTS: Lazy<Value> = Lazy::new(|| {
+    json!([
+        {
+            "title": "Shazam!",
+            "release_year": 2019,
+            "id": "287947",
+            // Three semantic properties:
+            // 1. magic, anything that reminds you of magic
+            // 2. authority, anything that inspires command
+            // 3. horror, anything that inspires fear or dread
+            "_vectors": { "manual": [0.8, 0.4, -0.5]},
+        },
+        {
+            "title": "Captain Marvel",
+            "release_year": 2019,
+            "id": "299537",
+            "_vectors": { "manual": [0.6, 0.8, -0.2] },
+        },
+        {
+            "title": "Escape Room",
+            "release_year": 2019,
+            "id": "522681",
+            "_vectors": { "manual": [0.1, 0.6, 0.8] },
+        },
+        {
+            "title": "How to Train Your Dragon: The Hidden World",
+            "release_year": 2019,
+            "id": "166428",
+            "_vectors": { "manual": [0.7, 0.7, -0.4] },
+        },
+        {
+            "title": "All Quiet on the Western Front",
+            "release_year": 1930,
+            "id": "143",
+            "_vectors": { "manual": [-0.5, 0.3, 0.85] },
+        }
+    ])
+});
+
+#[actix_rt::test]
+async fn basic() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    let (value, code) = server.set_features(json!({"vectorStore": true})).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(value, @r###"
+    {
+      "vectorStore": true,
+      "metrics": false,
+      "logsRoute": false
+    }
+    "###);
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    index
+        .similar(json!({"id": 143}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "Escape Room",
+                "release_year": 2019,
+                "id": "522681",
+                "_vectors": {
+                  "manual": [
+                    0.1,
+                    0.6,
+                    0.8
+                  ]
+                }
+              },
+              {
+                "title": "Captain Marvel",
+                "release_year": 2019,
+                "id": "299537",
+                "_vectors": {
+                  "manual": [
+                    0.6,
+                    0.8,
+                    -0.2
+                  ]
+                }
+              },
+              {
+                "title": "How to Train Your Dragon: The Hidden World",
+                "release_year": 2019,
+                "id": "166428",
+                "_vectors": {
+                  "manual": [
+                    0.7,
+                    0.7,
+                    -0.4
+                  ]
+                }
+              },
+              {
+                "title": "Shazam!",
+                "release_year": 2019,
+                "id": "287947",
+                "_vectors": {
+                  "manual": [
+                    0.8,
+                    0.4,
+                    -0.5
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+
+    index
+        .similar(json!({"id": "299537"}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "How to Train Your Dragon: The Hidden World",
+                "release_year": 2019,
+                "id": "166428",
+                "_vectors": {
+                  "manual": [
+                    0.7,
+                    0.7,
+                    -0.4
+                  ]
+                }
+              },
+              {
+                "title": "Shazam!",
+                "release_year": 2019,
+                "id": "287947",
+                "_vectors": {
+                  "manual": [
+                    0.8,
+                    0.4,
+                    -0.5
+                  ]
+                }
+              },
+              {
+                "title": "Escape Room",
+                "release_year": 2019,
+                "id": "522681",
+                "_vectors": {
+                  "manual": [
+                    0.1,
+                    0.6,
+                    0.8
+                  ]
+                }
+              },
+              {
+                "title": "All Quiet on the Western Front",
+                "release_year": 1930,
+                "id": "143",
+                "_vectors": {
+                  "manual": [
+                    -0.5,
+                    0.3,
+                    0.85
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn filter() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    let (value, code) = server.set_features(json!({"vectorStore": true})).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(value, @r###"
+    {
+      "vectorStore": true,
+      "metrics": false,
+      "logsRoute": false
+    }
+    "###);
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title", "release_year"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    index
+        .similar(json!({"id": 522681, "filter": "release_year = 2019"}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "Captain Marvel",
+                "release_year": 2019,
+                "id": "299537",
+                "_vectors": {
+                  "manual": [
+                    0.6,
+                    0.8,
+                    -0.2
+                  ]
+                }
+              },
+              {
+                "title": "How to Train Your Dragon: The Hidden World",
+                "release_year": 2019,
+                "id": "166428",
+                "_vectors": {
+                  "manual": [
+                    0.7,
+                    0.7,
+                    -0.4
+                  ]
+                }
+              },
+              {
+                "title": "Shazam!",
+                "release_year": 2019,
+                "id": "287947",
+                "_vectors": {
+                  "manual": [
+                    0.8,
+                    0.4,
+                    -0.5
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+
+    index
+        .similar(json!({"id": 522681, "filter": "release_year < 2000"}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "All Quiet on the Western Front",
+                "release_year": 1930,
+                "id": "143",
+                "_vectors": {
+                  "manual": [
+                    -0.5,
+                    0.3,
+                    0.85
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+}
+
+#[actix_rt::test]
+async fn limit_and_offset() {
+    let server = Server::new().await;
+    let index = server.index("test");
+    let (value, code) = server.set_features(json!({"vectorStore": true})).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(value, @r###"
+    {
+      "vectorStore": true,
+      "metrics": false,
+      "logsRoute": false
+    }
+    "###);
+
+    let (response, code) = index
+        .update_settings(json!({
+        "embedders": {
+            "manual": {
+                "source": "userProvided",
+                "dimensions": 3,
+            }
+        },
+        "filterableAttributes": ["title"]}))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    server.wait_task(response.uid()).await;
+
+    let documents = DOCUMENTS.clone();
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    index.wait_task(value.uid()).await;
+
+    index
+        .similar(json!({"id": 143, "limit": 1}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "Escape Room",
+                "release_year": 2019,
+                "id": "522681",
+                "_vectors": {
+                  "manual": [
+                    0.1,
+                    0.6,
+                    0.8
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+
+    index
+        .similar(json!({"id": 143, "limit": 1, "offset": 1}), |response, code| {
+            snapshot!(code, @"200 OK");
+            snapshot!(json_string!(response["hits"]), @r###"
+            [
+              {
+                "title": "Captain Marvel",
+                "release_year": 2019,
+                "id": "299537",
+                "_vectors": {
+                  "manual": [
+                    0.6,
+                    0.8,
+                    -0.2
+                  ]
+                }
+              }
+            ]
+            "###);
+        })
+        .await;
+}
--- a/meilisearch/tests/snapshot/mod.rs
+++ b/meilisearch/tests/snapshot/mod.rs
@ -1,6 +1,5 @@
 use std::time::Duration;

-use actix_rt::time::sleep;
 use meili_snap::{json_string, snapshot};
 use meilisearch::option::ScheduleSnapshot;
 use meilisearch::Opt;
@ -32,6 +31,7 @@ macro_rules! verify_snapshot {
 }

 #[actix_rt::test]
+#[cfg_attr(target_os = "windows", ignore)]
 async fn perform_snapshot() {
    let temp = tempfile::tempdir().unwrap();
    let snapshot_dir = tempfile::tempdir().unwrap();
@ -53,11 +53,29 @@ async fn perform_snapshot() {

    index.load_test_set().await;

-    server.index("test1").create(Some("prim")).await;
+    let (task, code) = server.index("test1").create(Some("prim")).await;
+    meili_snap::snapshot!(code, @"202 Accepted");

-    index.wait_task(2).await;
+    index.wait_task(task.uid()).await;

-    sleep(Duration::from_secs(2)).await;
+    // wait for the _next task_ to process, aka the snapshot that should be enqueued at some point
+
+    println!("waited for the next task to finish");
+    let now = std::time::Instant::now();
+    let next_task = task.uid() + 1;
+    loop {
+        let (value, code) = index.get_task(next_task).await;
+        dbg!(&value);
+        if code != 404 && value["status"].as_str() == Some("succeeded") {
+            break;
+        }
+
+        if now.elapsed() > Duration::from_secs(30) {
+            panic!("The snapshot didn't schedule in 30s even though it was supposed to be scheduled every 2s: {}",
+                serde_json::to_string_pretty(&value).unwrap()
+            );
+        }
+    }

    let temp = tempfile::tempdir().unwrap();

--- a/meilitool/src/main.rs
+++ b/meilitool/src/main.rs
@ -80,9 +80,7 @@ fn main() -> anyhow::Result<()> {
 /// Clears the task queue located at `db_path`.
 fn clear_task_queue(db_path: PathBuf) -> anyhow::Result<()> {
    let path = db_path.join("tasks");
-    let env = EnvOpenOptions::new()
-        .max_dbs(100)
-        .open(&path)
+    let env = unsafe { EnvOpenOptions::new().max_dbs(100).open(&path) }
        .with_context(|| format!("While trying to open {:?}", path.display()))?;

    eprintln!("Deleting tasks from the database...");
@ -193,9 +191,7 @@ fn export_a_dump(
        FileStore::new(db_path.join("update_files")).context("While opening the FileStore")?;

    let index_scheduler_path = db_path.join("tasks");
-    let env = EnvOpenOptions::new()
-        .max_dbs(100)
-        .open(&index_scheduler_path)
+    let env = unsafe { EnvOpenOptions::new().max_dbs(100).open(&index_scheduler_path) }
        .with_context(|| format!("While trying to open {:?}", index_scheduler_path.display()))?;

    eprintln!("Dumping the keys...");
--- a/milli/Cargo.toml
+++ b/milli/Cargo.toml
@ -30,7 +30,7 @@ grenad = { version = "0.4.6", default-features = false, features = [
    "rayon",
    "tempfile",
 ] }
-heed = { version = "0.20.0-alpha.9", default-features = false, features = [
+heed = { version = "0.20.1", default-features = false, features = [
    "serde-json",
    "serde-bincode",
    "read-txn-no-tls",
@ -67,9 +67,6 @@ filter-parser = { path = "../filter-parser" }
 # documents words self-join
 itertools = "0.11.0"

-# profiling
-puffin = "0.16.0"
-
 csv = "1.3.0"
 candle-core = { version = "0.4.1" }
 candle-transformers = { version = "0.4.1" }
@ -82,7 +79,7 @@ hf-hub = { git = "https://github.com/dureuill/hf-hub.git", branch = "rust_tls",
 ] }
 tiktoken-rs = "0.5.8"
 liquid = "0.26.4"
-arroy = "0.2.0"
+arroy = "0.3.1"
 rand = "0.8.5"
 tracing = "0.1.40"
 ureq = { version = "2.9.7", features = ["json"] }
--- a/milli/examples/search.rs
+++ b/milli/examples/search.rs
@ -48,8 +48,8 @@ fn main() -> Result<(), Box<dyn Error>> {

            let start = Instant::now();

-            let mut ctx = SearchContext::new(&index, &txn);
-            let universe = filtered_universe(&ctx, &None)?;
+            let mut ctx = SearchContext::new(&index, &txn)?;
+            let universe = filtered_universe(ctx.index, ctx.txn, &None)?;

            let docs = execute_search(
                &mut ctx,
--- a/milli/fuzz/.gitignore
+++ b/milli/fuzz/.gitignore
@ -0,0 +1,3 @@
+target
+corpus
+artifacts
--- a/milli/src/documents/mod.rs
+++ b/milli/src/documents/mod.rs
@ -12,7 +12,10 @@ use bimap::BiHashMap;
 pub use builder::DocumentsBatchBuilder;
 pub use enriched::{EnrichedDocument, EnrichedDocumentsBatchCursor, EnrichedDocumentsBatchReader};
 use obkv::KvReader;
-pub use primary_key::{DocumentIdExtractionError, FieldIdMapper, PrimaryKey, DEFAULT_PRIMARY_KEY};
+pub use primary_key::{
+    validate_document_id_value, DocumentIdExtractionError, FieldIdMapper, PrimaryKey,
+    DEFAULT_PRIMARY_KEY,
+};
 pub use reader::{DocumentsBatchCursor, DocumentsBatchCursorError, DocumentsBatchReader};
 use serde::{Deserialize, Serialize};

--- a/milli/src/documents/primary_key.rs
+++ b/milli/src/documents/primary_key.rs
@ -60,7 +60,7 @@ impl<'a> PrimaryKey<'a> {
                Some(document_id_bytes) => {
                    let document_id = serde_json::from_slice(document_id_bytes)
                        .map_err(InternalError::SerdeJson)?;
-                    match validate_document_id_value(document_id)? {
+                    match validate_document_id_value(document_id) {
                        Ok(document_id) => Ok(Ok(document_id)),
                        Err(user_error) => {
                            Ok(Err(DocumentIdExtractionError::InvalidDocumentId(user_error)))
@ -88,7 +88,7 @@ impl<'a> PrimaryKey<'a> {
                }

                match matching_documents_ids.pop() {
-                    Some(document_id) => match validate_document_id_value(document_id)? {
+                    Some(document_id) => match validate_document_id_value(document_id) {
                        Ok(document_id) => Ok(Ok(document_id)),
                        Err(user_error) => {
                            Ok(Err(DocumentIdExtractionError::InvalidDocumentId(user_error)))
@ -159,14 +159,14 @@ fn validate_document_id(document_id: &str) -> Option<&str> {
    }
 }

-pub fn validate_document_id_value(document_id: Value) -> Result<StdResult<String, UserError>> {
+pub fn validate_document_id_value(document_id: Value) -> StdResult<String, UserError> {
    match document_id {
        Value::String(string) => match validate_document_id(&string) {
-            Some(s) if s.len() == string.len() => Ok(Ok(string)),
-            Some(s) => Ok(Ok(s.to_string())),
-            None => Ok(Err(UserError::InvalidDocumentId { document_id: Value::String(string) })),
+            Some(s) if s.len() == string.len() => Ok(string),
+            Some(s) => Ok(s.to_string()),
+            None => Err(UserError::InvalidDocumentId { document_id: Value::String(string) }),
        },
-        Value::Number(number) if number.is_i64() => Ok(Ok(number.to_string())),
-        content => Ok(Err(UserError::InvalidDocumentId { document_id: content })),
+        Value::Number(number) if number.is_i64() => Ok(number.to_string()),
+        content => Err(UserError::InvalidDocumentId { document_id: content }),
    }
 }
--- a/milli/src/error.rs
+++ b/milli/src/error.rs
@ -32,6 +32,8 @@ pub enum InternalError {
    DatabaseClosing,
    #[error("Missing {} in the {db_name} database.", key.unwrap_or("key"))]
    DatabaseMissingEntry { db_name: &'static str, key: Option<&'static str> },
+    #[error("Missing {key} in the fieldids weights mapping.")]
+    FieldidsWeightsMapMissingEntry { key: FieldId },
    #[error(transparent)]
    FieldIdMapMissingEntry(#[from] FieldIdMapMissingEntry),
    #[error("Missing {key} in the field id mapping.")]
@ -46,8 +48,6 @@ pub enum InternalError {
    GrenadInvalidFormatVersion,
    #[error("Invalid merge while processing {process}")]
    IndexingMergingKeys { process: &'static str },
-    #[error("{}", HeedError::InvalidDatabaseTyping)]
-    InvalidDatabaseTyping,
    #[error(transparent)]
    RayonThreadPool(#[from] ThreadPoolBuildError),
    #[error(transparent)]
@ -117,10 +117,8 @@ only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and undersco
    InvalidGeoField(#[from] GeoError),
    #[error("Invalid vector dimensions: expected: `{}`, found: `{}`.", .expected, .found)]
    InvalidVectorDimensions { expected: usize, found: usize },
-    #[error("The `_vectors.{subfield}` field in the document with id: `{document_id}` is not an array. Was expecting an array of floats or an array of arrays of floats but instead got `{value}`.")]
-    InvalidVectorsType { document_id: Value, value: Value, subfield: String },
    #[error("The `_vectors` field in the document with id: `{document_id}` is not an object. Was expecting an object with a key for each embedder with manually provided vectors, but instead got `{value}`")]
-    InvalidVectorsMapType { document_id: Value, value: Value },
+    InvalidVectorsMapType { document_id: String, value: Value },
    #[error("{0}")]
    InvalidFilter(String),
    #[error("Invalid type for filter subexpression: expected: {}, found: {1}.", .0.join(", "))]
@ -427,7 +425,6 @@ impl From<HeedError> for Error {
            // TODO use the encoding
            HeedError::Encoding(_) => InternalError(Serialization(Encoding { db_name: None })),
            HeedError::Decoding(_) => InternalError(Serialization(Decoding { db_name: None })),
-            HeedError::InvalidDatabaseTyping => InternalError(InvalidDatabaseTyping),
            HeedError::DatabaseClosing => InternalError(DatabaseClosing),
            HeedError::BadOpenOptions { .. } => UserError(InvalidLmdbOpenOptions),
        }
--- a/milli/src/fieldids_weights_map.rs
+++ b/milli/src/fieldids_weights_map.rs
@ -0,0 +1,48 @@
+//! The fieldids weights map is in charge of storing linking the searchable fields with their weights.
+
+use std::collections::HashMap;
+
+use serde::{Deserialize, Serialize};
+
+use crate::{FieldId, FieldsIdsMap, Weight};
+
+#[derive(Debug, Default, Serialize, Deserialize)]
+pub struct FieldidsWeightsMap {
+    map: HashMap<FieldId, Weight>,
+}
+
+impl FieldidsWeightsMap {
+    /// Insert a field id -> weigth into the map.
+    /// If the map did not have this key present, `None` is returned.
+    /// If the map did have this key present, the value is updated, and the old value is returned.
+    pub fn insert(&mut self, fid: FieldId, weight: Weight) -> Option<Weight> {
+        self.map.insert(fid, weight)
+    }
+
+    /// Create the map from the fields ids maps.
+    /// Should only be called in the case there are NO searchable attributes.
+    /// All the fields will be inserted in the order of the fields ids map with a weight of 0.
+    pub fn from_field_id_map_without_searchable(fid_map: &FieldsIdsMap) -> Self {
+        FieldidsWeightsMap { map: fid_map.ids().map(|fid| (fid, 0)).collect() }
+    }
+
+    /// Removes a field id from the map, returning the associated weight previously in the map.
+    pub fn remove(&mut self, fid: FieldId) -> Option<Weight> {
+        self.map.remove(&fid)
+    }
+
+    /// Returns weight corresponding to the key.
+    pub fn weight(&self, fid: FieldId) -> Option<Weight> {
+        self.map.get(&fid).copied()
+    }
+
+    /// Returns highest weight contained in the map if any.
+    pub fn max_weight(&self) -> Option<Weight> {
+        self.map.values().copied().max()
+    }
+
+    /// Return an iterator visiting all field ids in arbitrary order.
+    pub fn ids(&self) -> impl Iterator<Item = FieldId> + '_ {
+        self.map.keys().copied()
+    }
+}
--- a/milli/src/heed_codec/roaring_bitmap/cbo_roaring_bitmap_codec.rs
+++ b/milli/src/heed_codec/roaring_bitmap/cbo_roaring_bitmap_codec.rs
@ -195,7 +195,7 @@ mod tests {
    fn merge_cbo_roaring_bitmaps() {
        let mut buffer = Vec::new();

-        let small_data = vec![
+        let small_data = [
            RoaringBitmap::from_sorted_iter(1..4).unwrap(),
            RoaringBitmap::from_sorted_iter(2..5).unwrap(),
            RoaringBitmap::from_sorted_iter(4..6).unwrap(),
@ -209,7 +209,7 @@ mod tests {
        let expected = RoaringBitmap::from_sorted_iter(1..6).unwrap();
        assert_eq!(bitmap, expected);

-        let medium_data = vec![
+        let medium_data = [
            RoaringBitmap::from_sorted_iter(1..4).unwrap(),
            RoaringBitmap::from_sorted_iter(2..5).unwrap(),
            RoaringBitmap::from_sorted_iter(4..8).unwrap(),
--- a/milli/src/index.rs
+++ b/milli/src/index.rs
@ -1,5 +1,6 @@
 use std::borrow::Cow;
 use std::collections::{BTreeMap, BTreeSet, HashMap, HashSet};
+use std::convert::TryInto;
 use std::fs::File;
 use std::path::Path;

@ -8,6 +9,7 @@ use heed::types::*;
 use heed::{CompactionOption, Database, RoTxn, RwTxn, Unspecified};
 use roaring::RoaringBitmap;
 use rstar::RTree;
+use serde::Serialize;
 use time::OffsetDateTime;

 use crate::documents::PrimaryKey;
@ -22,11 +24,12 @@ use crate::heed_codec::{
 };
 use crate::order_by_map::OrderByMap;
 use crate::proximity::ProximityPrecision;
-use crate::vector::EmbeddingConfig;
+use crate::vector::{Embedding, EmbeddingConfig};
 use crate::{
    default_criteria, CboRoaringBitmapCodec, Criterion, DocumentId, ExternalDocumentsIds,
-    FacetDistribution, FieldDistribution, FieldId, FieldIdWordCountCodec, GeoPoint, ObkvCodec,
-    Result, RoaringBitmapCodec, RoaringBitmapLenCodec, Search, U8StrStrCodec, BEU16, BEU32, BEU64,
+    FacetDistribution, FieldDistribution, FieldId, FieldIdMapMissingEntry, FieldIdWordCountCodec,
+    FieldidsWeightsMap, GeoPoint, ObkvCodec, Result, RoaringBitmapCodec, RoaringBitmapLenCodec,
+    Search, U8StrStrCodec, Weight, BEU16, BEU32, BEU64,
 };

 pub const DEFAULT_MIN_WORD_LEN_ONE_TYPO: u8 = 5;
@ -42,6 +45,7 @@ pub mod main_key {
    pub const SORTABLE_FIELDS_KEY: &str = "sortable-fields";
    pub const FIELD_DISTRIBUTION_KEY: &str = "fields-distribution";
    pub const FIELDS_IDS_MAP_KEY: &str = "fields-ids-map";
+    pub const FIELDIDS_WEIGHTS_MAP_KEY: &str = "fieldids-weights-map";
    pub const GEO_FACETED_DOCUMENTS_IDS_KEY: &str = "geo-faceted-documents-ids";
    pub const GEO_RTREE_KEY: &str = "geo-rtree";
    pub const PRIMARY_KEY_KEY: &str = "primary-key";
@ -181,7 +185,7 @@ impl Index {

        options.max_dbs(25);

-        let env = options.open(path)?;
+        let env = unsafe { options.open(path) }?;
        let mut wtxn = env.write_txn()?;
        let main = env.database_options().name(MAIN).create(&mut wtxn)?;
        let word_docids = env.create_database(&mut wtxn, Some(WORD_DOCIDS))?;
@ -291,6 +295,11 @@ impl Index {
        self.env.read_txn()
    }

+    /// Create a static read transaction to be able to read the index without keeping a reference to it.
+    pub fn static_read_txn(&self) -> heed::Result<RoTxn<'static>> {
+        self.env.clone().static_read_txn()
+    }
+
    /// Returns the canonicalized path where the heed `Env` of this `Index` lives.
    pub fn path(&self) -> &Path {
        self.env.path()
@ -316,6 +325,87 @@ impl Index {
        self.env.info().map_size
    }

+    pub fn advanced_stats(&self, rtxn: &heed::RoTxn) -> Result<AdvancedStats> {
+        use db_name::*;
+
+        let mut dbs = BTreeMap::new();
+        dbs.insert(WORD_DOCIDS, advanced_database_stats(rtxn, self.word_docids)?);
+        dbs.insert(
+            WORD_PAIR_PROXIMITY_DOCIDS,
+            advanced_database_stats(rtxn, self.word_pair_proximity_docids)?,
+        );
+        dbs.insert(WORD_PREFIX_DOCIDS, advanced_database_stats(rtxn, self.word_prefix_docids)?);
+        dbs.insert(WORD_FIELD_ID_DOCIDS, advanced_database_stats(rtxn, self.word_fid_docids)?);
+        dbs.insert(WORD_POSITION_DOCIDS, advanced_database_stats(rtxn, self.word_position_docids)?);
+        dbs.insert(DOCUMENTS, advanced_database_stats_no_bitmap(rtxn, self.documents)?);
+
+        fn advanced_database_stats<KC>(
+            rtxn: &heed::RoTxn,
+            db: Database<KC, CboRoaringBitmapCodec>,
+        ) -> Result<AdvancedDatabaseStats> {
+            let db = db.remap_key_type::<Bytes>().lazily_decode_data();
+
+            let mut entries_count = 0;
+            let mut total_bitmap_size = 0;
+            let mut total_bitmap_len = 0;
+            let mut total_key_size = 0;
+
+            for result in db.iter(rtxn)? {
+                let (bytes_key, lazy_value) = result?;
+                entries_count += 1;
+                total_bitmap_size += lazy_value.remap::<Bytes>().decode().unwrap().len();
+                let bitmap = lazy_value.decode().map_err(heed::Error::Decoding)?;
+                total_bitmap_len += bitmap.len();
+                total_key_size += bytes_key.len();
+            }
+
+            Ok(AdvancedDatabaseStats {
+                entries_count,
+                average_bitmap_len: Some(total_bitmap_len as f64 / entries_count as f64),
+                median_bitmap_len: None,
+                average_value_size: Some(total_bitmap_size as f64 / entries_count as f64),
+                median_value_size: None,
+                average_key_size: Some(total_key_size as f64 / entries_count as f64),
+                median_key_size: None,
+            })
+        }
+
+        fn advanced_database_stats_no_bitmap<KC, DC>(
+            rtxn: &heed::RoTxn,
+            db: Database<KC, DC>,
+        ) -> Result<AdvancedDatabaseStats> {
+            let db = db.remap_types::<Bytes, Bytes>();
+
+            let mut entries_count = 0;
+            let mut total_value_size = 0;
+            let mut total_key_size = 0;
+
+            for result in db.iter(rtxn)? {
+                let (bytes_key, bytes_value) = result?;
+                entries_count += 1;
+                total_value_size += bytes_value.len();
+                total_key_size += bytes_key.len();
+            }
+
+            Ok(AdvancedDatabaseStats {
+                entries_count,
+                average_bitmap_len: None,
+                median_bitmap_len: None,
+                average_value_size: Some(total_value_size as f64 / entries_count as f64),
+                median_value_size: None,
+                average_key_size: Some(total_key_size as f64 / entries_count as f64),
+                median_key_size: None,
+            })
+        }
+
+        Ok(AdvancedStats {
+            map_size: self.map_size(),
+            non_free_pages_size: self.on_disk_size()?,
+            on_disk_size: self.on_disk_size()?,
+            databases: dbs,
+        })
+    }
+
    pub fn copy_to_file<P: AsRef<Path>>(&self, path: P, option: CompactionOption) -> Result<File> {
        self.env.copy_to_file(path, option).map_err(Into::into)
    }
@ -414,6 +504,65 @@ impl Index {
            .unwrap_or_default())
    }

+    /* fieldids weights map */
+    // This maps the fields ids to their weights.
+    // Their weights is defined by the ordering of the searchable attributes.
+
+    /// Writes the fieldids weights map which associates the field ids to their weights
+    pub(crate) fn put_fieldids_weights_map(
+        &self,
+        wtxn: &mut RwTxn,
+        map: &FieldidsWeightsMap,
+    ) -> heed::Result<()> {
+        self.main.remap_types::<Str, SerdeJson<_>>().put(
+            wtxn,
+            main_key::FIELDIDS_WEIGHTS_MAP_KEY,
+            map,
+        )
+    }
+
+    /// Get the fieldids weights map which associates the field ids to their weights
+    pub fn fieldids_weights_map(&self, rtxn: &RoTxn) -> heed::Result<FieldidsWeightsMap> {
+        self.main
+            .remap_types::<Str, SerdeJson<_>>()
+            .get(rtxn, main_key::FIELDIDS_WEIGHTS_MAP_KEY)?
+            .map(Ok)
+            .unwrap_or_else(|| {
+                Ok(FieldidsWeightsMap::from_field_id_map_without_searchable(
+                    &self.fields_ids_map(rtxn)?,
+                ))
+            })
+    }
+
+    /// Delete the fieldsids weights map
+    pub fn delete_fieldids_weights_map(&self, wtxn: &mut RwTxn) -> heed::Result<bool> {
+        self.main.remap_key_type::<Str>().delete(wtxn, main_key::FIELDIDS_WEIGHTS_MAP_KEY)
+    }
+
+    pub fn searchable_fields_and_weights<'a>(
+        &self,
+        rtxn: &'a RoTxn,
+    ) -> Result<Vec<(Cow<'a, str>, FieldId, Weight)>> {
+        let fid_map = self.fields_ids_map(rtxn)?;
+        let weight_map = self.fieldids_weights_map(rtxn)?;
+        let searchable = self.searchable_fields(rtxn)?;
+
+        searchable
+            .into_iter()
+            .map(|field| -> Result<_> {
+                let fid = fid_map.id(&field).ok_or_else(|| FieldIdMapMissingEntry::FieldName {
+                    field_name: field.to_string(),
+                    process: "searchable_fields_and_weights",
+                })?;
+                let weight = weight_map
+                    .weight(fid)
+                    .ok_or(InternalError::FieldidsWeightsMapMissingEntry { key: fid })?;
+
+                Ok((field, fid, weight))
+            })
+            .collect()
+    }
+
    /* geo rtree */

    /// Writes the provided `rtree` which associates coordinates to documents ids.
@ -578,33 +727,42 @@ impl Index {
        wtxn: &mut RwTxn,
        user_fields: &[&str],
        fields_ids_map: &FieldsIdsMap,
-    ) -> heed::Result<()> {
+    ) -> Result<()> {
        // We can write the user defined searchable fields as-is.
        self.put_user_defined_searchable_fields(wtxn, user_fields)?;

+        let mut weights = FieldidsWeightsMap::default();
+
        // Now we generate the real searchable fields:
        // 1. Take the user defined searchable fields as-is to keep the priority defined by the attributes criterion.
        // 2. Iterate over the user defined searchable fields.
        // 3. If a user defined field is a subset of a field defined in the fields_ids_map
-        // (ie doggo.name is a subset of doggo) then we push it at the end of the fields.
-        let mut real_fields = user_fields.to_vec();
+        // (ie doggo.name is a subset of doggo) right after doggo and with the same weight.
+        let mut real_fields = Vec::new();

-        for field_from_map in fields_ids_map.names() {
-            for user_field in user_fields {
+        for (id, field_from_map) in fields_ids_map.iter() {
+            for (weight, user_field) in user_fields.iter().enumerate() {
                if crate::is_faceted_by(field_from_map, user_field)
-                    && !user_fields.contains(&field_from_map)
+                    && !real_fields.contains(&field_from_map)
                {
                    real_fields.push(field_from_map);
+
+                    let weight: u16 =
+                        weight.try_into().map_err(|_| UserError::AttributeLimitReached)?;
+                    weights.insert(id, weight);
                }
            }
        }

-        self.put_searchable_fields(wtxn, &real_fields)
+        self.put_searchable_fields(wtxn, &real_fields)?;
+        self.put_fieldids_weights_map(wtxn, &weights)?;
+        Ok(())
    }

    pub(crate) fn delete_all_searchable_fields(&self, wtxn: &mut RwTxn) -> heed::Result<bool> {
        let did_delete_searchable = self.delete_searchable_fields(wtxn)?;
        let did_delete_user_defined = self.delete_user_defined_searchable_fields(wtxn)?;
+        self.delete_fieldids_weights_map(wtxn)?;
        Ok(did_delete_searchable || did_delete_user_defined)
    }

@ -623,28 +781,31 @@ impl Index {
    }

    /// Returns the searchable fields, those are the fields that are indexed,
-    /// if the searchable fields aren't there it means that **all** the fields are indexed.
-    pub fn searchable_fields<'t>(&self, rtxn: &'t RoTxn) -> heed::Result<Option<Vec<&'t str>>> {
+    pub fn searchable_fields<'t>(&self, rtxn: &'t RoTxn) -> heed::Result<Vec<Cow<'t, str>>> {
        self.main
            .remap_types::<Str, SerdeBincode<Vec<&'t str>>>()
-            .get(rtxn, main_key::SEARCHABLE_FIELDS_KEY)
+            .get(rtxn, main_key::SEARCHABLE_FIELDS_KEY)?
+            .map(|fields| Ok(fields.into_iter().map(Cow::Borrowed).collect()))
+            .unwrap_or_else(|| {
+                Ok(self
+                    .fields_ids_map(rtxn)?
+                    .names()
+                    .map(|field| Cow::Owned(field.to_string()))
+                    .collect())
+            })
    }

    /// Identical to `searchable_fields`, but returns the ids instead.
-    pub fn searchable_fields_ids(&self, rtxn: &RoTxn) -> Result<Option<Vec<FieldId>>> {
-        match self.searchable_fields(rtxn)? {
-            Some(fields) => {
-                let fields_ids_map = self.fields_ids_map(rtxn)?;
-                let mut fields_ids = Vec::new();
-                for name in fields {
-                    if let Some(field_id) = fields_ids_map.id(name) {
-                        fields_ids.push(field_id);
-                    }
-                }
-                Ok(Some(fields_ids))
+    pub fn searchable_fields_ids(&self, rtxn: &RoTxn) -> Result<Vec<FieldId>> {
+        let fields = self.searchable_fields(rtxn)?;
+        let fields_ids_map = self.fields_ids_map(rtxn)?;
+        let mut fields_ids = Vec::new();
+        for name in fields {
+            if let Some(field_id) = fields_ids_map.id(&name) {
+                fields_ids.push(field_id);
            }
-            None => Ok(None),
        }
+        Ok(fields_ids)
    }

    /// Writes the searchable fields, when this list is specified, only these are indexed.
@ -1516,6 +1677,22 @@ impl Index {
            .unwrap_or_default())
    }

+    pub fn arroy_readers<'a>(
+        &'a self,
+        rtxn: &'a RoTxn<'a>,
+        embedder_id: u8,
+    ) -> impl Iterator<Item = Result<arroy::Reader<arroy::distances::Angular>>> + 'a {
+        crate::vector::arroy_db_range_for_embedder(embedder_id).map_while(move |k| {
+            arroy::Reader::open(rtxn, k, self.vector_arroy)
+                .map(Some)
+                .or_else(|e| match e {
+                    arroy::Error::MissingMetadata => Ok(None),
+                    e => Err(e.into()),
+                })
+                .transpose()
+        })
+    }
+
    pub(crate) fn put_search_cutoff(&self, wtxn: &mut RwTxn<'_>, cutoff: u64) -> heed::Result<()> {
        self.main.remap_types::<Str, BEU64>().put(wtxn, main_key::SEARCH_CUTOFF, &cutoff)
    }
@ -1527,6 +1704,74 @@ impl Index {
    pub(crate) fn delete_search_cutoff(&self, wtxn: &mut RwTxn<'_>) -> heed::Result<bool> {
        self.main.remap_key_type::<Str>().delete(wtxn, main_key::SEARCH_CUTOFF)
    }
+
+    pub fn embeddings(
+        &self,
+        rtxn: &RoTxn<'_>,
+        docid: DocumentId,
+    ) -> Result<BTreeMap<String, Vec<Embedding>>> {
+        let mut res = BTreeMap::new();
+        for row in self.embedder_category_id.iter(rtxn)? {
+            let (embedder_name, embedder_id) = row?;
+            let embedder_id = (embedder_id as u16) << 8;
+            let mut embeddings = Vec::new();
+            'vectors: for i in 0..=u8::MAX {
+                let reader = arroy::Reader::open(rtxn, embedder_id | (i as u16), self.vector_arroy)
+                    .map(Some)
+                    .or_else(|e| match e {
+                        arroy::Error::MissingMetadata => Ok(None),
+                        e => Err(e),
+                    })
+                    .transpose();
+
+                let Some(reader) = reader else {
+                    break 'vectors;
+                };
+
+                let embedding = reader?.item_vector(rtxn, docid)?;
+                if let Some(embedding) = embedding {
+                    embeddings.push(embedding)
+                } else {
+                    break 'vectors;
+                }
+            }
+
+            if !embeddings.is_empty() {
+                res.insert(embedder_name.to_owned(), embeddings);
+            }
+        }
+        Ok(res)
+    }
+}
+
+#[derive(Clone, Debug, Serialize)]
+pub struct AdvancedStats {
+    /// Size of the data memory map.
+    map_size: usize,
+    /// Returns the size used by all the databases in the environment without the free pages.
+    non_free_pages_size: u64,
+    /// The size of the data file on disk.
+    on_disk_size: u64,
+    /// Databases advanced stats.
+    databases: BTreeMap<&'static str, AdvancedDatabaseStats>,
+}
+
+#[derive(Clone, Debug, Serialize)]
+pub struct AdvancedDatabaseStats {
+    /// The number of entries in this database.
+    entries_count: usize,
+    /// The average number of entries in the bitmaps of this database.
+    average_bitmap_len: Option<f64>,
+    /// The median number of entries in the bitmaps of this database.
+    median_bitmap_len: Option<f64>,
+    /// The average size of values of this database.
+    average_value_size: Option<f64>,
+    /// The median size of values of this database.
+    median_value_size: Option<f64>,
+    /// The average size of keys of this database.
+    average_key_size: Option<f64>,
+    /// The mediane size of keys of this database.
+    median_key_size: Option<f64>,
 }

 #[cfg(test)]
@ -1710,10 +1955,14 @@ pub(crate) mod tests {
            ]))
            .unwrap();

-        db_snap!(index, field_distribution, 1);
+        db_snap!(index, field_distribution, @r###"
+        age              1      |
+        id               2      |
+        name             2      |
+        "###);

        db_snap!(index, word_docids,
-            @r###"
+        @r###"
        1                [0, ]
        2                [1, ]
        20               [1, ]
@ -1722,18 +1971,6 @@ pub(crate) mod tests {
        "###
        );

-        db_snap!(index, field_distribution);
-
-        db_snap!(index, field_distribution,
-            @r###"
-        age              1      |
-        id               2      |
-        name             2      |
-        "###
-        );
-
-        // snapshot_index!(&index, "1", include: "^field_distribution$");
-
        // we add all the documents a second time. we are supposed to get the same
        // field_distribution in the end
        index
@ -1820,7 +2057,7 @@ pub(crate) mod tests {
        // ensure we get the right real searchable fields + user defined searchable fields
        let rtxn = index.read_txn().unwrap();

-        let real = index.searchable_fields(&rtxn).unwrap().unwrap();
+        let real = index.searchable_fields(&rtxn).unwrap();
        assert_eq!(real, &["doggo", "name", "doggo.name", "doggo.age"]);

        let user_defined = index.user_defined_searchable_fields(&rtxn).unwrap().unwrap();
@ -1840,7 +2077,7 @@ pub(crate) mod tests {
        // ensure we get the right real searchable fields + user defined searchable fields
        let rtxn = index.read_txn().unwrap();

-        let real = index.searchable_fields(&rtxn).unwrap().unwrap();
+        let real = index.searchable_fields(&rtxn).unwrap();
        assert_eq!(real, &["doggo", "name"]);
        let user_defined = index.user_defined_searchable_fields(&rtxn).unwrap().unwrap();
        assert_eq!(user_defined, &["doggo", "name"]);
@ -1856,7 +2093,7 @@ pub(crate) mod tests {
        // ensure we get the right real searchable fields + user defined searchable fields
        let rtxn = index.read_txn().unwrap();

-        let real = index.searchable_fields(&rtxn).unwrap().unwrap();
+        let real = index.searchable_fields(&rtxn).unwrap();
        assert_eq!(real, &["doggo", "name", "doggo.name", "doggo.age"]);

        let user_defined = index.user_defined_searchable_fields(&rtxn).unwrap().unwrap();
@ -2395,6 +2632,14 @@ pub(crate) mod tests {
        11                       0
        4                        1
        "###);
+        db_snap!(index, fields_ids_map, @r###"
+        0   primary_key      |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["primary_key"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        "###);

        index
            .add_documents(documents!([
@ -2410,6 +2655,16 @@ pub(crate) mod tests {
        11                       0
        4                        1
        "###);
+        db_snap!(index, fields_ids_map, @r###"
+        0   primary_key      |
+        1   a                |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["primary_key", "a"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        1   0   |
+        "###);

        index.delete_documents(Default::default());

@ -2420,6 +2675,16 @@ pub(crate) mod tests {
        11                       0
        4                        1
        "###);
+        db_snap!(index, fields_ids_map, @r###"
+        0   primary_key      |
+        1   a                |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["primary_key", "a"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        1   0   |
+        "###);

        index
            .add_documents(documents!([
@ -2435,6 +2700,16 @@ pub(crate) mod tests {
        11                       0
        4                        1
        "###);
+        db_snap!(index, fields_ids_map, @r###"
+        0   primary_key      |
+        1   a                |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["primary_key", "a"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        1   0   |
+        "###);

        let rtxn = index.read_txn().unwrap();
        let search = Search::new(&rtxn, &index);
@ -2520,4 +2795,104 @@ pub(crate) mod tests {

        db_snap!(index, geo_faceted_documents_ids); // ensure that no documents were inserted
    }
+
+    #[test]
+    fn swapping_searchable_attributes() {
+        // See https://github.com/meilisearch/meilisearch/issues/4484
+
+        let index = TempIndex::new();
+
+        index
+            .update_settings(|settings| {
+                settings.set_searchable_fields(vec![S("name")]);
+                settings.set_filterable_fields(HashSet::from([S("age")]));
+            })
+            .unwrap();
+
+        index
+            .add_documents(documents!({ "id": 1, "name": "Many", "age": 28, "realName": "Maxime" }))
+            .unwrap();
+        db_snap!(index, fields_ids_map, @r###"
+        0   name             |
+        1   id               |
+        2   age              |
+        3   realName         |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["name"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        "###);
+
+        index
+            .update_settings(|settings| {
+                settings.set_searchable_fields(vec![S("name"), S("realName")]);
+                settings.set_filterable_fields(HashSet::from([S("age")]));
+            })
+            .unwrap();
+
+        // The order of the field id map shouldn't change
+        db_snap!(index, fields_ids_map, @r###"
+        0   name             |
+        1   id               |
+        2   age              |
+        3   realName         |
+        "###);
+        db_snap!(index, searchable_fields, @r###"["name", "realName"]"###);
+        db_snap!(index, fieldids_weights_map, @r###"
+        fid weight
+        0   0   |
+        3   1   |
+        "###);
+    }
+
+    #[test]
+    fn attribute_weights_after_swapping_searchable_attributes() {
+        // See https://github.com/meilisearch/meilisearch/issues/4484
+
+        let index = TempIndex::new();
+
+        index
+            .update_settings(|settings| {
+                settings.set_searchable_fields(vec![S("name"), S("beverage")]);
+            })
+            .unwrap();
+
+        index
+            .add_documents(documents!([
+                { "id": 0, "name": "kefir", "beverage": "water" },
+                { "id": 1, "name": "tamo",  "beverage": "kefir" }
+            ]))
+            .unwrap();
+
+        let rtxn = index.read_txn().unwrap();
+        let mut search = index.search(&rtxn);
+        let results = search.query("kefir").execute().unwrap();
+
+        // We should find kefir the dog first
+        insta::assert_debug_snapshot!(results.documents_ids, @r###"
+        [
+            0,
+            1,
+        ]
+        "###);
+
+        index
+            .update_settings(|settings| {
+                settings.set_searchable_fields(vec![S("beverage"), S("name")]);
+            })
+            .unwrap();
+
+        let rtxn = index.read_txn().unwrap();
+        let mut search = index.search(&rtxn);
+        let results = search.query("kefir").execute().unwrap();
+
+        // We should find tamo first
+        insta::assert_debug_snapshot!(results.documents_ids, @r###"
+        [
+            1,
+            0,
+        ]
+        "###);
+    }
 }
--- a/milli/src/lib.rs
+++ b/milli/src/lib.rs
@ -28,6 +28,7 @@ pub mod vector;
 #[cfg(test)]
 #[macro_use]
 pub mod snapshot_tests;
+mod fieldids_weights_map;

 use std::collections::{BTreeMap, HashMap};
 use std::convert::{TryFrom, TryInto};
@ -52,6 +53,7 @@ pub use self::error::{
    Error, FieldIdMapMissingEntry, InternalError, SerializationError, UserError,
 };
 pub use self::external_documents_ids::ExternalDocumentsIds;
+pub use self::fieldids_weights_map::FieldidsWeightsMap;
 pub use self::fields_ids_map::FieldsIdsMap;
 pub use self::heed_codec::{
    BEU16StrCodec, BEU32StrCodec, BoRoaringBitmapCodec, BoRoaringBitmapLenCodec,
@ -61,6 +63,7 @@ pub use self::heed_codec::{
 };
 pub use self::index::Index;
 pub use self::search::facet::{FacetValueHit, SearchForFacetValues};
+pub use self::search::similar::Similar;
 pub use self::search::{
    FacetDistribution, Filter, FormatOptions, MatchBounds, MatcherBuilder, MatchingWords, OrderBy,
    Search, SearchResult, SemanticSearch, TermsMatchingStrategy, DEFAULT_VALUES_PER_FACET,
@ -77,6 +80,7 @@ pub type FastMap4<K, V> = HashMap<K, V, BuildHasherDefault<FxHasher32>>;
 pub type FastMap8<K, V> = HashMap<K, V, BuildHasherDefault<FxHasher64>>;
 pub type FieldDistribution = BTreeMap<String, u64>;
 pub type FieldId = u16;
+pub type Weight = u16;
 pub type Object = serde_json::Map<String, serde_json::Value>;
 pub type Position = u32;
 pub type RelativePosition = u16;
@ -351,43 +355,13 @@ pub fn is_faceted(field: &str, faceted_fields: impl IntoIterator<Item = impl AsR
 /// assert!(!is_faceted_by("animaux.chien", "animaux.chie"));
 /// ```
 pub fn is_faceted_by(field: &str, facet: &str) -> bool {
-    field.starts_with(facet)
-        && field[facet.len()..].chars().next().map(|c| c == '.').unwrap_or(true)
+    field.starts_with(facet) && field[facet.len()..].chars().next().map_or(true, |c| c == '.')
 }

 pub fn normalize_facet(original: &str) -> String {
    CompatibilityDecompositionNormalizer.normalize_str(original.trim()).to_lowercase()
 }

-/// Represents either a vector or an array of multiple vectors.
-#[derive(serde::Serialize, serde::Deserialize, Debug)]
-#[serde(transparent)]
-pub struct VectorOrArrayOfVectors {
-    #[serde(with = "either::serde_untagged_optional")]
-    inner: Option<either::Either<Vec<f32>, Vec<Vec<f32>>>>,
-}
-
-impl VectorOrArrayOfVectors {
-    pub fn into_array_of_vectors(self) -> Option<Vec<Vec<f32>>> {
-        match self.inner? {
-            either::Either::Left(vector) => Some(vec![vector]),
-            either::Either::Right(vectors) => Some(vectors),
-        }
-    }
-}
-
-/// Normalize a vector by dividing the dimensions by the length of it.
-pub fn normalize_vector(mut vector: Vec<f32>) -> Vec<f32> {
-    let squared: f32 = vector.iter().map(|x| x * x).sum();
-    let length = squared.sqrt();
-    if length <= f32::EPSILON {
-        vector
-    } else {
-        vector.iter_mut().for_each(|x| *x /= length);
-        vector
-    }
-}
-
 #[cfg(test)]
 mod tests {
    use serde_json::json;
--- a/milli/src/search/mod.rs
+++ b/milli/src/search/mod.rs
@ -24,6 +24,7 @@ pub mod facet;
 mod fst_utils;
 pub mod hybrid;
 pub mod new;
+pub mod similar;

 #[derive(Debug, Clone)]
 pub struct SemanticSearch {
@ -147,21 +148,21 @@ impl<'a> Search<'a> {

    pub fn execute_for_candidates(&self, has_vector_search: bool) -> Result<RoaringBitmap> {
        if has_vector_search {
-            let ctx = SearchContext::new(self.index, self.rtxn);
-            filtered_universe(&ctx, &self.filter)
+            let ctx = SearchContext::new(self.index, self.rtxn)?;
+            filtered_universe(ctx.index, ctx.txn, &self.filter)
        } else {
            Ok(self.execute()?.candidates)
        }
    }

    pub fn execute(&self) -> Result<SearchResult> {
-        let mut ctx = SearchContext::new(self.index, self.rtxn);
+        let mut ctx = SearchContext::new(self.index, self.rtxn)?;

        if let Some(searchable_attributes) = self.searchable_attributes {
-            ctx.searchable_attributes(searchable_attributes)?;
+            ctx.attributes_to_search_on(searchable_attributes)?;
        }

-        let universe = filtered_universe(&ctx, &self.filter)?;
+        let universe = filtered_universe(ctx.index, ctx.txn, &self.filter)?;
        let PartialSearchResult {
            located_query_terms,
            candidates,
@ -276,6 +277,8 @@ pub enum TermsMatchingStrategy {
    Last,
    // all words are mandatory
    All,
+    // remove more frequent word first
+    Frequency,
 }

 impl Default for TermsMatchingStrategy {
--- a/milli/src/search/new/bucket_sort.rs
+++ b/milli/src/search/new/bucket_sort.rs
@ -101,7 +101,7 @@ pub fn bucket_sort<'ctx, Q: RankingRuleQueryTrait>(

    let mut ranking_rule_universes: Vec<RoaringBitmap> =
        vec![RoaringBitmap::default(); ranking_rules_len];
-    ranking_rule_universes[0] = universe.clone();
+    ranking_rule_universes[0].clone_from(universe);
    let mut cur_ranking_rule_index = 0;

    /// Finish iterating over the current ranking rule, yielding
@ -232,7 +232,7 @@ pub fn bucket_sort<'ctx, Q: RankingRuleQueryTrait>(
        }

        cur_ranking_rule_index += 1;
-        ranking_rule_universes[cur_ranking_rule_index] = next_bucket.candidates.clone();
+        ranking_rule_universes[cur_ranking_rule_index].clone_from(&next_bucket.candidates);
        logger.start_iteration_ranking_rule(
            cur_ranking_rule_index,
            ranking_rules[cur_ranking_rule_index].as_ref(),
--- a/milli/src/search/new/db_cache.rs
+++ b/milli/src/search/new/db_cache.rs
@ -163,7 +163,7 @@ impl<'ctx> SearchContext<'ctx> {
            Some(restricted_fids) => {
                let interned = self.word_interner.get(word).as_str();
                let keys: Vec<_> =
-                    restricted_fids.tolerant.iter().map(|fid| (interned, *fid)).collect();
+                    restricted_fids.tolerant.iter().map(|(fid, _)| (interned, *fid)).collect();

                DatabaseCache::get_value_from_keys::<_, _, CboRoaringBitmapCodec>(
                    self.txn,
@ -192,7 +192,7 @@ impl<'ctx> SearchContext<'ctx> {
            Some(restricted_fids) => {
                let interned = self.word_interner.get(word).as_str();
                let keys: Vec<_> =
-                    restricted_fids.exact.iter().map(|fid| (interned, *fid)).collect();
+                    restricted_fids.exact.iter().map(|(fid, _)| (interned, *fid)).collect();

                DatabaseCache::get_value_from_keys::<_, _, CboRoaringBitmapCodec>(
                    self.txn,
@ -242,7 +242,7 @@ impl<'ctx> SearchContext<'ctx> {
            Some(restricted_fids) => {
                let interned = self.word_interner.get(prefix).as_str();
                let keys: Vec<_> =
-                    restricted_fids.tolerant.iter().map(|fid| (interned, *fid)).collect();
+                    restricted_fids.tolerant.iter().map(|(fid, _)| (interned, *fid)).collect();

                DatabaseCache::get_value_from_keys::<_, _, CboRoaringBitmapCodec>(
                    self.txn,
@ -271,7 +271,7 @@ impl<'ctx> SearchContext<'ctx> {
            Some(restricted_fids) => {
                let interned = self.word_interner.get(prefix).as_str();
                let keys: Vec<_> =
-                    restricted_fids.exact.iter().map(|fid| (interned, *fid)).collect();
+                    restricted_fids.exact.iter().map(|(fid, _)| (interned, *fid)).collect();

                DatabaseCache::get_value_from_keys::<_, _, CboRoaringBitmapCodec>(
                    self.txn,
@ -315,11 +315,7 @@ impl<'ctx> SearchContext<'ctx> {
                        .map_err(heed::Error::Decoding)?
                } else {
                    // Compute the distance at the attribute level and store it in the cache.
-                    let fids = if let Some(fids) = self.index.searchable_fields_ids(self.txn)? {
-                        fids
-                    } else {
-                        self.index.fields_ids_map(self.txn)?.ids().collect()
-                    };
+                    let fids = self.index.searchable_fields_ids(self.txn)?;
                    let mut docids = RoaringBitmap::new();
                    for fid in fids {
                        // for each field, intersect left word bitmap and right word bitmap,
@ -408,11 +404,7 @@ impl<'ctx> SearchContext<'ctx> {
            let prefix_docids = match proximity_precision {
                ProximityPrecision::ByAttribute => {
                    // Compute the distance at the attribute level and store it in the cache.
-                    let fids = if let Some(fids) = self.index.searchable_fields_ids(self.txn)? {
-                        fids
-                    } else {
-                        self.index.fields_ids_map(self.txn)?.ids().collect()
-                    };
+                    let fids = self.index.searchable_fields_ids(self.txn)?;
                    let mut prefix_docids = RoaringBitmap::new();
                    // for each field, intersect left word bitmap and right word bitmap,
                    // then merge the result in a global bitmap before storing it in the cache.
--- a/milli/src/search/new/exact_attribute.rs
+++ b/milli/src/search/new/exact_attribute.rs
@ -184,13 +184,7 @@ impl State {
            return Ok(State::Empty(query_graph.clone()));
        }

-        let searchable_fields_ids = {
-            if let Some(fids) = ctx.index.searchable_fields_ids(ctx.txn)? {
-                fids
-            } else {
-                ctx.index.fields_ids_map(ctx.txn)?.ids().collect()
-            }
-        };
+        let searchable_fields_ids = ctx.index.searchable_fields_ids(ctx.txn)?;

        let mut candidates_per_attribute = Vec::with_capacity(searchable_fields_ids.len());
        // then check that there exists at least one attribute that has all of the terms
--- a/milli/src/search/new/graph_based_ranking_rule.rs
+++ b/milli/src/search/new/graph_based_ranking_rule.rs
@ -164,6 +164,21 @@ impl<'ctx, G: RankingRuleGraphTrait> RankingRule<'ctx, QueryGraph> for GraphBase
                    }
                    costs
                }
+                TermsMatchingStrategy::Frequency => {
+                    let removal_order =
+                        query_graph.removal_order_for_terms_matching_strategy_frequency(ctx)?;
+                    let mut forbidden_nodes =
+                        SmallBitmap::for_interned_values_in(&query_graph.nodes);
+                    let mut costs = query_graph.nodes.map(|_| None);
+                    // FIXME: this works because only words uses termsmatchingstrategy at the moment.
+                    for ns in removal_order {
+                        for n in ns.iter() {
+                            *costs.get_mut(n) = Some((1, forbidden_nodes.clone()));
+                        }
+                        forbidden_nodes.union(&ns);
+                    }
+                    costs
+                }
                TermsMatchingStrategy::All => query_graph.nodes.map(|_| None),
            }
        } else {
--- a/milli/src/search/new/matches/matching_words.rs
+++ b/milli/src/search/new/matches/matching_words.rs
@ -258,7 +258,7 @@ pub(crate) mod tests {
    fn matching_words() {
        let temp_index = temp_index_with_documents();
        let rtxn = temp_index.read_txn().unwrap();
-        let mut ctx = SearchContext::new(&temp_index, &rtxn);
+        let mut ctx = SearchContext::new(&temp_index, &rtxn).unwrap();
        let mut builder = TokenizerBuilder::default();
        let tokenizer = builder.build();
        let tokens = tokenizer.tokenize("split this world");
--- a/milli/src/search/new/matches/mod.rs
+++ b/milli/src/search/new/matches/mod.rs
@ -506,8 +506,8 @@ mod tests {

    impl<'a> MatcherBuilder<'a> {
        fn new_test(rtxn: &'a heed::RoTxn, index: &'a TempIndex, query: &str) -> Self {
-            let mut ctx = SearchContext::new(index, rtxn);
-            let universe = filtered_universe(&ctx, &None).unwrap();
+            let mut ctx = SearchContext::new(index, rtxn).unwrap();
+            let universe = filtered_universe(ctx.index, ctx.txn, &None).unwrap();
            let crate::search::PartialSearchResult { located_query_terms, .. } = execute_search(
                &mut ctx,
                Some(query),
--- a/milli/src/search/new/mod.rs
+++ b/milli/src/search/new/mod.rs
@ -49,13 +49,12 @@ pub use self::geo_sort::Strategy as GeoSortStrategy;
 use self::graph_based_ranking_rule::Words;
 use self::interner::Interned;
 use self::vector_sort::VectorSort;
-use crate::error::FieldIdMapMissingEntry;
 use crate::score_details::{ScoreDetails, ScoringStrategy};
 use crate::search::new::distinct::apply_distinct_rule;
 use crate::vector::Embedder;
 use crate::{
    AscDesc, DocumentId, FieldId, Filter, Index, Member, Result, TermsMatchingStrategy, TimeBudget,
-    UserError,
+    UserError, Weight,
 };

 /// A structure used throughout the execution of a search query.
@ -71,8 +70,21 @@ pub struct SearchContext<'ctx> {
 }

 impl<'ctx> SearchContext<'ctx> {
-    pub fn new(index: &'ctx Index, txn: &'ctx RoTxn<'ctx>) -> Self {
-        Self {
+    pub fn new(index: &'ctx Index, txn: &'ctx RoTxn<'ctx>) -> Result<Self> {
+        let searchable_fids = index.searchable_fields_and_weights(txn)?;
+        let exact_attributes_ids = index.exact_attributes_ids(txn)?;
+
+        let mut exact = Vec::new();
+        let mut tolerant = Vec::new();
+        for (_name, fid, weight) in searchable_fids {
+            if exact_attributes_ids.contains(&fid) {
+                exact.push((fid, weight));
+            } else {
+                tolerant.push((fid, weight));
+            }
+        }
+
+        Ok(Self {
            index,
            txn,
            db_cache: <_>::default(),
@ -81,42 +93,39 @@ impl<'ctx> SearchContext<'ctx> {
            term_interner: <_>::default(),
            phrase_docids: <_>::default(),
            restricted_fids: None,
-        }
+        })
    }

-    pub fn searchable_attributes(&mut self, searchable_attributes: &'ctx [String]) -> Result<()> {
-        let fids_map = self.index.fields_ids_map(self.txn)?;
-        let searchable_names = self.index.searchable_fields(self.txn)?;
+    pub fn attributes_to_search_on(
+        &mut self,
+        attributes_to_search_on: &'ctx [String],
+    ) -> Result<()> {
+        let user_defined_searchable = self.index.user_defined_searchable_fields(self.txn)?;
+        let searchable_fields_weights = self.index.searchable_fields_and_weights(self.txn)?;
        let exact_attributes_ids = self.index.exact_attributes_ids(self.txn)?;

+        let mut wildcard = false;
+
        let mut restricted_fids = RestrictedFids::default();
-        let mut contains_wildcard = false;
-        for field_name in searchable_attributes {
+        for field_name in attributes_to_search_on {
            if field_name == "*" {
-                contains_wildcard = true;
+                wildcard = true;
+                // we cannot early exit as we want to returns error in case of unknown fields
                continue;
            }
-            let searchable_contains_name =
-                searchable_names.as_ref().map(|sn| sn.iter().any(|name| name == field_name));
-            let fid = match (fids_map.id(field_name), searchable_contains_name) {
+            let searchable_weight =
+                searchable_fields_weights.iter().find(|(name, _, _)| name == field_name);
+            let (fid, weight) = match searchable_weight {
                // The Field id exist and the field is searchable
-                (Some(fid), Some(true)) | (Some(fid), None) => fid,
-                // The field is searchable but the Field id doesn't exist => Internal Error
-                (None, Some(true)) => {
-                    return Err(FieldIdMapMissingEntry::FieldName {
-                        field_name: field_name.to_string(),
-                        process: "search",
-                    }
-                    .into())
-                }
-                // The field is not searchable, but the searchableAttributes are set to * => ignore field
-                (None, None) => continue,
+                Some((_name, fid, weight)) => (*fid, *weight),
+                // The field is not searchable but the user didn't define any searchable attributes
+                None if user_defined_searchable.is_none() => continue,
                // The field is not searchable => User error
-                (_fid, Some(false)) => {
-                    let (valid_fields, hidden_fields) = match searchable_names {
-                        Some(sn) => self.index.remove_hidden_fields(self.txn, sn)?,
-                        None => self.index.remove_hidden_fields(self.txn, fids_map.names())?,
-                    };
+                None => {
+                    let (valid_fields, hidden_fields) = self.index.remove_hidden_fields(
+                        self.txn,
+                        searchable_fields_weights.iter().map(|(name, _, _)| name),
+                    )?;

                    let field = field_name.to_string();
                    return Err(UserError::InvalidSearchableAttribute {
@ -129,13 +138,17 @@ impl<'ctx> SearchContext<'ctx> {
            };

            if exact_attributes_ids.contains(&fid) {
-                restricted_fids.exact.push(fid);
+                restricted_fids.exact.push((fid, weight));
            } else {
-                restricted_fids.tolerant.push(fid);
+                restricted_fids.tolerant.push((fid, weight));
            };
        }

-        self.restricted_fids = (!contains_wildcard).then_some(restricted_fids);
+        if wildcard {
+            self.restricted_fids = None;
+        } else {
+            self.restricted_fids = Some(restricted_fids);
+        }

        Ok(())
    }
@ -158,13 +171,13 @@ impl Word {

 #[derive(Debug, Clone, Default)]
 pub struct RestrictedFids {
-    pub tolerant: Vec<FieldId>,
-    pub exact: Vec<FieldId>,
+    pub tolerant: Vec<(FieldId, Weight)>,
+    pub exact: Vec<(FieldId, Weight)>,
 }

 impl RestrictedFids {
    pub fn contains(&self, fid: &FieldId) -> bool {
-        self.tolerant.contains(fid) || self.exact.contains(fid)
+        self.tolerant.iter().any(|(id, _)| id == fid) || self.exact.iter().any(|(id, _)| id == fid)
    }
 }

@ -184,6 +197,11 @@ fn resolve_maximally_reduced_query_graph(
            .iter()
            .flat_map(|x| x.iter())
            .collect(),
+        TermsMatchingStrategy::Frequency => query_graph
+            .removal_order_for_terms_matching_strategy_frequency(ctx)?
+            .iter()
+            .flat_map(|x| x.iter())
+            .collect(),
        TermsMatchingStrategy::All => vec![],
    };
    graph.remove_nodes_keep_edges(&nodes_to_remove);
@ -530,11 +548,15 @@ fn resolve_sort_criteria<'ctx, Query: RankingRuleQueryTrait>(
    Ok(())
 }

-pub fn filtered_universe(ctx: &SearchContext, filters: &Option<Filter>) -> Result<RoaringBitmap> {
+pub fn filtered_universe(
+    index: &Index,
+    txn: &RoTxn<'_>,
+    filters: &Option<Filter>,
+) -> Result<RoaringBitmap> {
    Ok(if let Some(filters) = filters {
-        filters.evaluate(ctx.txn, ctx.index)?
+        filters.evaluate(txn, index)?
    } else {
-        ctx.index.documents_ids(ctx.txn)?
+        index.documents_ids(txn)?
    })
 }

--- a/milli/src/search/new/query_graph.rs
+++ b/milli/src/search/new/query_graph.rs
@ -1,8 +1,9 @@
-use std::cmp::Ordering;
+use std::cmp::{Ordering, Reverse};
 use std::collections::BTreeMap;
 use std::hash::{Hash, Hasher};

 use fxhash::{FxHashMap, FxHasher};
+use roaring::RoaringBitmap;

 use super::interner::{FixedSizeInterner, Interned};
 use super::query_term::{
@ -11,6 +12,7 @@ use super::query_term::{
 use super::small_bitmap::SmallBitmap;
 use super::SearchContext;
 use crate::search::new::interner::Interner;
+use crate::search::new::resolve_query_graph::compute_query_term_subset_docids;
 use crate::Result;

 /// A node of the [`QueryGraph`].
@ -290,6 +292,49 @@ impl QueryGraph {
        }
    }

+    pub fn removal_order_for_terms_matching_strategy_frequency(
+        &self,
+        ctx: &mut SearchContext,
+    ) -> Result<Vec<SmallBitmap<QueryNode>>> {
+        // lookup frequency for each term
+        let mut term_with_frequency: Vec<(u8, u64)> = {
+            let mut term_docids: BTreeMap<u8, RoaringBitmap> = Default::default();
+            for (_, node) in self.nodes.iter() {
+                match &node.data {
+                    QueryNodeData::Term(t) => {
+                        let docids = compute_query_term_subset_docids(ctx, &t.term_subset)?;
+                        for id in t.term_ids.clone() {
+                            term_docids
+                                .entry(id)
+                                .and_modify(|curr| *curr |= &docids)
+                                .or_insert_with(|| docids.clone());
+                        }
+                    }
+                    QueryNodeData::Deleted | QueryNodeData::Start | QueryNodeData::End => continue,
+                }
+            }
+            term_docids
+                .into_iter()
+                .map(|(idx, docids)| match docids.len() {
+                    0 => (idx, u64::max_value()),
+                    frequency => (idx, frequency),
+                })
+                .collect()
+        };
+        term_with_frequency.sort_by_key(|(_, frequency)| Reverse(*frequency));
+        let mut term_weight = BTreeMap::new();
+        let mut weight: u16 = 1;
+        let mut peekable = term_with_frequency.into_iter().peekable();
+        while let Some((idx, frequency)) = peekable.next() {
+            term_weight.insert(idx, weight);
+            if peekable.peek().map_or(false, |(_, f)| frequency != *f) {
+                weight += 1;
+            }
+        }
+        let cost_of_term_idx = move |term_idx: u8| *term_weight.get(&term_idx).unwrap();
+        Ok(self.removal_order_for_terms_matching_strategy(ctx, cost_of_term_idx))
+    }
+
    pub fn removal_order_for_terms_matching_strategy_last(
        &self,
        ctx: &SearchContext,
@ -315,10 +360,19 @@ impl QueryGraph {
        if first_term_idx >= last_term_idx {
            return vec![];
        }
+
        let cost_of_term_idx = |term_idx: u8| {
            let rank = 1 + last_term_idx - term_idx;
            rank as u16
        };
+        self.removal_order_for_terms_matching_strategy(ctx, cost_of_term_idx)
+    }
+
+    pub fn removal_order_for_terms_matching_strategy(
+        &self,
+        ctx: &SearchContext,
+        order: impl Fn(u8) -> u16,
+    ) -> Vec<SmallBitmap<QueryNode>> {
        let mut nodes_to_remove = BTreeMap::<u16, SmallBitmap<QueryNode>>::new();
        let mut at_least_one_mandatory_term = false;
        for (node_id, node) in self.nodes.iter() {
@ -329,7 +383,7 @@ impl QueryGraph {
            }
            let mut cost = 0;
            for id in t.term_ids.clone() {
-                cost = std::cmp::max(cost, cost_of_term_idx(id));
+                cost = std::cmp::max(cost, order(id));
            }
            nodes_to_remove
                .entry(cost)
--- a/milli/src/search/new/query_term/parse_query.rs
+++ b/milli/src/search/new/query_term/parse_query.rs
@ -366,7 +366,7 @@ mod tests {
        let tokens = tokenizer.tokenize(".");
        let index = temp_index_with_documents();
        let rtxn = index.read_txn()?;
-        let mut ctx = SearchContext::new(&index, &rtxn);
+        let mut ctx = SearchContext::new(&index, &rtxn)?;
        // panics with `attempt to add with overflow` before <https://github.com/meilisearch/meilisearch/issues/3785>
        let ExtractedTokens { query_terms, .. } =
            located_query_terms_from_tokens(&mut ctx, tokens, None)?;
--- a/milli/src/search/new/ranking_rule_graph/fid/mod.rs
+++ b/milli/src/search/new/ranking_rule_graph/fid/mod.rs
@ -7,12 +7,12 @@ use crate::search::new::interner::{DedupInterner, Interned};
 use crate::search::new::query_term::LocatedQueryTermSubset;
 use crate::search::new::resolve_query_graph::compute_query_term_subset_docids_within_field_id;
 use crate::search::new::SearchContext;
-use crate::Result;
+use crate::{FieldId, InternalError, Result};

 #[derive(Clone, PartialEq, Eq, Hash)]
 pub struct FidCondition {
    term: LocatedQueryTermSubset,
-    fid: u16,
+    fid: Option<FieldId>,
 }

 pub enum FidGraph {}
@ -26,13 +26,15 @@ impl RankingRuleGraphTrait for FidGraph {
        universe: &RoaringBitmap,
    ) -> Result<ComputedCondition> {
        let FidCondition { term, .. } = condition;
-        // maybe compute_query_term_subset_docids_within_field_id should accept a universe as argument
-        let mut docids = compute_query_term_subset_docids_within_field_id(
-            ctx,
-            &term.term_subset,
-            condition.fid,
-        )?;
-        docids &= universe;
+
+        let docids = if let Some(fid) = condition.fid {
+            // maybe compute_query_term_subset_docids_within_field_id should accept a universe as argument
+            let docids =
+                compute_query_term_subset_docids_within_field_id(ctx, &term.term_subset, fid)?;
+            docids & universe
+        } else {
+            RoaringBitmap::new()
+        };

        Ok(ComputedCondition {
            docids,
@ -68,34 +70,29 @@ impl RankingRuleGraphTrait for FidGraph {
            all_fields.extend(fields);
        }

+        let weights_map = ctx.index.fieldids_weights_map(ctx.txn)?;
+
        let mut edges = vec![];
        for fid in all_fields.iter().copied() {
+            let weight = weights_map
+                .weight(fid)
+                .ok_or(InternalError::FieldidsWeightsMapMissingEntry { key: fid })?;
            edges.push((
-                fid as u32 * term.term_ids.len() as u32,
-                conditions_interner.insert(FidCondition { term: term.clone(), fid }),
+                weight as u32 * term.term_ids.len() as u32,
+                conditions_interner.insert(FidCondition { term: term.clone(), fid: Some(fid) }),
            ));
        }

        // always lookup the max_fid if we don't already and add an artificial condition for max scoring
-        let max_fid: Option<u16> = {
-            if let Some(max_fid) = ctx
-                .index
-                .searchable_fields_ids(ctx.txn)?
-                .map(|field_ids| field_ids.into_iter().max())
-            {
-                max_fid
-            } else {
-                ctx.index.fields_ids_map(ctx.txn)?.ids().max()
-            }
-        };
+        let max_weight: Option<u16> = weights_map.max_weight();

-        if let Some(max_fid) = max_fid {
-            if !all_fields.contains(&max_fid) {
+        if let Some(max_weight) = max_weight {
+            if !all_fields.contains(&max_weight) {
                edges.push((
-                    max_fid as u32 * term.term_ids.len() as u32, // TODO improve the fid score i.e. fid^10.
+                    max_weight as u32 * term.term_ids.len() as u32, // TODO improve the fid score i.e. fid^10.
                    conditions_interner.insert(FidCondition {
                        term: term.clone(), // TODO remove this ugly clone
-                        fid: max_fid,
+                        fid: None,
                    }),
                ));
            }
--- a/milli/src/search/new/tests/attribute_fid.rs
+++ b/milli/src/search/new/tests/attribute_fid.rs
@ -1,5 +1,5 @@
 use crate::index::tests::TempIndex;
-use crate::{Criterion, Search, SearchResult, TermsMatchingStrategy};
+use crate::{db_snap, Criterion, Search, SearchResult, TermsMatchingStrategy};

 fn create_index() -> TempIndex {
    let index = TempIndex::new();
@ -131,6 +131,19 @@ fn test_attribute_fid_simple() {
 #[test]
 fn test_attribute_fid_ngrams() {
    let index = create_index();
+    db_snap!(index, fields_ids_map, @r###"
+    0   id               |
+    1   title            |
+    2   description      |
+    3   plot             |
+    "###);
+    db_snap!(index, searchable_fields, @r###"["title", "description", "plot"]"###);
+    db_snap!(index, fieldids_weights_map, @r###"
+    fid weight
+    1   0   |
+    2   1   |
+    3   2   |
+    "###);

    let txn = index.read_txn().unwrap();

--- a/milli/src/search/new/tests/snapshots/millisearchnewtestsattribute_fid__attribute_fid_ngrams-4.snap
+++ b/milli/src/search/new/tests/snapshots/millisearchnewtestsattribute_fid__attribute_fid_ngrams-4.snap
@ -0,0 +1,244 @@
+---
+source: milli/src/search/new/tests/attribute_fid.rs
+expression: "format!(\"{document_ids_scores:#?}\")"
+---
+[
+    (
+        2,
+        [
+            Fid(
+                Rank {
+                    rank: 19,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 91,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        6,
+        [
+            Fid(
+                Rank {
+                    rank: 15,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 81,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        5,
+        [
+            Fid(
+                Rank {
+                    rank: 14,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 79,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        4,
+        [
+            Fid(
+                Rank {
+                    rank: 13,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 77,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        3,
+        [
+            Fid(
+                Rank {
+                    rank: 12,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 83,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        9,
+        [
+            Fid(
+                Rank {
+                    rank: 11,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 75,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        8,
+        [
+            Fid(
+                Rank {
+                    rank: 10,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 79,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        7,
+        [
+            Fid(
+                Rank {
+                    rank: 10,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 73,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        11,
+        [
+            Fid(
+                Rank {
+                    rank: 7,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 77,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        10,
+        [
+            Fid(
+                Rank {
+                    rank: 6,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 81,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        13,
+        [
+            Fid(
+                Rank {
+                    rank: 6,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 81,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        12,
+        [
+            Fid(
+                Rank {
+                    rank: 6,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 78,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        14,
+        [
+            Fid(
+                Rank {
+                    rank: 5,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 75,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+    (
+        0,
+        [
+            Fid(
+                Rank {
+                    rank: 1,
+                    max_rank: 19,
+                },
+            ),
+            Position(
+                Rank {
+                    rank: 91,
+                    max_rank: 91,
+                },
+            ),
+        ],
+    ),
+]
--- a/milli/src/search/new/vector_sort.rs
+++ b/milli/src/search/new/vector_sort.rs
@ -49,19 +49,8 @@ impl<Q: RankingRuleQueryTrait> VectorSort<Q> {
        ctx: &mut SearchContext<'_>,
        vector_candidates: &RoaringBitmap,
    ) -> Result<()> {
-        let writer_index = (self.embedder_index as u16) << 8;
-        let readers: std::result::Result<Vec<_>, _> = (0..=u8::MAX)
-            .map_while(|k| {
-                arroy::Reader::open(ctx.txn, writer_index | (k as u16), ctx.index.vector_arroy)
-                    .map(Some)
-                    .or_else(|e| match e {
-                        arroy::Error::MissingMetadata => Ok(None),
-                        e => Err(e),
-                    })
-                    .transpose()
-            })
-            .collect();
-
+        let readers: std::result::Result<Vec<_>, _> =
+            ctx.index.arroy_readers(ctx.txn, self.embedder_index).collect();
        let readers = readers?;

        let target = &self.target;
--- a/milli/src/search/similar.rs
+++ b/milli/src/search/similar.rs
@ -0,0 +1,111 @@
+use std::sync::Arc;
+
+use ordered_float::OrderedFloat;
+use roaring::RoaringBitmap;
+
+use crate::score_details::{self, ScoreDetails};
+use crate::vector::Embedder;
+use crate::{filtered_universe, DocumentId, Filter, Index, Result, SearchResult};
+
+pub struct Similar<'a> {
+    id: DocumentId,
+    // this should be linked to the String in the query
+    filter: Option<Filter<'a>>,
+    offset: usize,
+    limit: usize,
+    rtxn: &'a heed::RoTxn<'a>,
+    index: &'a Index,
+    embedder_name: String,
+    embedder: Arc<Embedder>,
+}
+
+impl<'a> Similar<'a> {
+    pub fn new(
+        id: DocumentId,
+        offset: usize,
+        limit: usize,
+        index: &'a Index,
+        rtxn: &'a heed::RoTxn<'a>,
+        embedder_name: String,
+        embedder: Arc<Embedder>,
+    ) -> Self {
+        Self { id, filter: None, offset, limit, rtxn, index, embedder_name, embedder }
+    }
+
+    pub fn filter(&mut self, filter: Filter<'a>) -> &mut Self {
+        self.filter = Some(filter);
+        self
+    }
+
+    pub fn execute(&self) -> Result<SearchResult> {
+        let universe = filtered_universe(self.index, self.rtxn, &self.filter)?;
+
+        let embedder_index =
+            self.index
+                .embedder_category_id
+                .get(self.rtxn, &self.embedder_name)?
+                .ok_or_else(|| crate::UserError::InvalidEmbedder(self.embedder_name.to_owned()))?;
+
+        let readers: std::result::Result<Vec<_>, _> =
+            self.index.arroy_readers(self.rtxn, embedder_index).collect();
+
+        let readers = readers?;
+
+        let mut results = Vec::new();
+
+        for reader in readers.iter() {
+            let nns_by_item = reader.nns_by_item(
+                self.rtxn,
+                self.id,
+                self.limit + self.offset + 1,
+                None,
+                Some(&universe),
+            )?;
+            if let Some(mut nns_by_item) = nns_by_item {
+                results.append(&mut nns_by_item);
+            } else {
+                break;
+            }
+        }
+
+        results.sort_unstable_by_key(|(_, distance)| OrderedFloat(*distance));
+
+        let mut documents_ids = Vec::with_capacity(self.limit);
+        let mut document_scores = Vec::with_capacity(self.limit);
+        // list of documents we've already seen, so that we don't return the same document multiple times.
+        // initialized to the target document, that we never want to return.
+        let mut documents_seen = RoaringBitmap::new();
+        documents_seen.insert(self.id);
+
+        for (docid, distance) in results
+            .into_iter()
+            // skip documents we've already seen & mark that we saw the current document
+            .filter(|(docid, _)| documents_seen.insert(*docid))
+            .skip(self.offset)
+            // take **after** filter and skip so that we get exactly limit elements if available
+            .take(self.limit)
+        {
+            documents_ids.push(docid);
+
+            let score = 1.0 - distance;
+            let score = self
+                .embedder
+                .distribution()
+                .map(|distribution| distribution.shift(score))
+                .unwrap_or(score);
+
+            let score = ScoreDetails::Vector(score_details::Vector { similarity: Some(score) });
+
+            document_scores.push(vec![score]);
+        }
+
+        Ok(SearchResult {
+            matching_words: Default::default(),
+            candidates: universe,
+            documents_ids,
+            document_scores,
+            degraded: false,
+            used_negative_operator: false,
+        })
+    }
+}
--- a/milli/src/snapshot_tests.rs
+++ b/milli/src/snapshot_tests.rs
@ -308,6 +308,25 @@ pub fn snap_fields_ids_map(index: &Index) -> String {
    }
    snap
 }
+pub fn snap_fieldids_weights_map(index: &Index) -> String {
+    let rtxn = index.read_txn().unwrap();
+    let weights_map = index.fieldids_weights_map(&rtxn).unwrap();
+
+    let mut snap = String::new();
+    writeln!(&mut snap, "fid weight").unwrap();
+    let mut field_ids: Vec<_> = weights_map.ids().collect();
+    field_ids.sort();
+    for field_id in field_ids {
+        let weight = weights_map.weight(field_id).unwrap();
+        writeln!(&mut snap, "{field_id:<3} {weight:<3} |").unwrap();
+    }
+    snap
+}
+pub fn snap_searchable_fields(index: &Index) -> String {
+    let rtxn = index.read_txn().unwrap();
+    let searchable_fields = index.searchable_fields(&rtxn).unwrap();
+    format!("{searchable_fields:?}")
+}
 pub fn snap_geo_faceted_documents_ids(index: &Index) -> String {
    let rtxn = index.read_txn().unwrap();
    let geo_faceted_documents_ids = index.geo_faceted_documents_ids(&rtxn).unwrap();
@ -469,6 +488,12 @@ macro_rules! full_snap_of_db {
    ($index:ident, fields_ids_map) => {{
        $crate::snapshot_tests::snap_fields_ids_map(&$index)
    }};
+    ($index:ident, fieldids_weights_map) => {{
+        $crate::snapshot_tests::snap_fieldids_weights_map(&$index)
+    }};
+    ($index:ident, searchable_fields) => {{
+        $crate::snapshot_tests::snap_searchable_fields(&$index)
+    }};
    ($index:ident, geo_faceted_documents_ids) => {{
        $crate::snapshot_tests::snap_geo_faceted_documents_ids(&$index)
    }};
--- a/milli/src/update/clear_documents.rs
+++ b/milli/src/update/clear_documents.rs
@ -21,8 +21,6 @@ impl<'t, 'i> ClearDocuments<'t, 'i> {
        name = "clear_documents"
    )]
    pub fn execute(self) -> Result<u64> {
-        puffin::profile_function!();
-
        self.index.set_updated_at(self.wtxn, &OffsetDateTime::now_utc())?;
        let Index {
            env: _env,
--- a/milli/src/update/facet/mod.rs
+++ b/milli/src/update/facet/mod.rs
@ -379,7 +379,7 @@ pub(crate) mod test_helpers {
            let mut options = heed::EnvOpenOptions::new();
            let options = options.map_size(4096 * 4 * 1000 * 100);
            let tempdir = tempfile::TempDir::new().unwrap();
-            let env = options.open(tempdir.path()).unwrap();
+            let env = unsafe { options.open(tempdir.path()) }.unwrap();
            let mut wtxn = env.write_txn().unwrap();
            let content = env.create_database(&mut wtxn, None).unwrap();
            wtxn.commit().unwrap();
--- a/milli/src/update/index_documents/enrich.rs
+++ b/milli/src/update/index_documents/enrich.rs
@ -29,8 +29,6 @@ pub fn enrich_documents_batch<R: Read + Seek>(
    autogenerate_docids: bool,
    reader: DocumentsBatchReader<R>,
 ) -> Result<StdResult<EnrichedDocumentsBatchReader<R>, UserError>> {
-    puffin::profile_function!();
-
    let (mut cursor, mut documents_batch_index) = reader.into_cursor_and_fields_index();

    let mut external_ids = tempfile::tempfile().map(BufWriter::new).map(grenad::Writer::new)?;
--- a/milli/src/update/index_documents/extract/extract_docid_word_positions.rs
+++ b/milli/src/update/index_documents/extract/extract_docid_word_positions.rs
@ -29,8 +29,6 @@ pub fn extract_docid_word_positions<R: io::Read + io::Seek>(
    settings_diff: &InnerIndexSettingsDiff,
    max_positions_per_attributes: Option<u32>,
 ) -> Result<(grenad::Reader<BufReader<File>>, ScriptLanguageDocidsMap)> {
-    puffin::profile_function!();
-
    let max_positions_per_attributes = max_positions_per_attributes
        .map_or(MAX_POSITION_PER_ATTRIBUTE, |max| max.min(MAX_POSITION_PER_ATTRIBUTE));
    let max_memory = indexer.max_memory_by_thread();
@ -186,7 +184,7 @@ fn searchable_fields_changed(
 ) -> bool {
    let searchable_fields = &settings_diff.new.searchable_fields_ids;
    for (field_id, field_bytes) in obkv.iter() {
-        if searchable_fields.as_ref().map_or(true, |sf| sf.contains(&field_id)) {
+        if searchable_fields.contains(&field_id) {
            let del_add = KvReaderDelAdd::new(field_bytes);
            match (del_add.get(DelAdd::Deletion), del_add.get(DelAdd::Addition)) {
                // if both fields are None, check the next field.
@ -298,7 +296,7 @@ fn lang_safe_tokens_from_document<'a>(
 /// Extract words mapped with their positions of a document.
 fn tokens_from_document<'a>(
    obkv: &KvReader<FieldId>,
-    searchable_fields: &Option<Vec<FieldId>>,
+    searchable_fields: &[FieldId],
    tokenizer: &Tokenizer,
    max_positions_per_attributes: u32,
    del_add: DelAdd,
@ -309,7 +307,7 @@ fn tokens_from_document<'a>(
    let mut document_writer = KvWriterU16::new(&mut buffers.obkv_buffer);
    for (field_id, field_bytes) in obkv.iter() {
        // if field is searchable.
-        if searchable_fields.as_ref().map_or(true, |sf| sf.contains(&field_id)) {
+        if searchable_fields.as_ref().contains(&field_id) {
            // extract deletion or addition only.
            if let Some(field_bytes) = KvReaderDelAdd::new(field_bytes).get(del_add) {
                // parse json.
--- a/milli/src/update/index_documents/extract/extract_facet_number_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_facet_number_docids.rs
@ -23,8 +23,6 @@ pub fn extract_facet_number_docids<R: io::Read + io::Seek>(
    indexer: GrenadParameters,
    _settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();

    let mut facet_number_docids_sorter = create_sorter(
--- a/milli/src/update/index_documents/extract/extract_facet_string_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_facet_string_docids.rs
@ -28,8 +28,6 @@ pub fn extract_facet_string_docids<R: io::Read + io::Seek>(
    indexer: GrenadParameters,
    _settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<(grenad::Reader<BufReader<File>>, grenad::Reader<BufReader<File>>)> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();
    let options = NormalizerOption { lossy: true, ..Default::default() };

--- a/milli/src/update/index_documents/extract/extract_fid_docid_facet_values.rs
+++ b/milli/src/update/index_documents/extract/extract_fid_docid_facet_values.rs
@ -45,10 +45,7 @@ pub fn extract_fid_docid_facet_values<R: io::Read + io::Seek>(
    obkv_documents: grenad::Reader<R>,
    indexer: GrenadParameters,
    settings_diff: &InnerIndexSettingsDiff,
-    geo_fields_ids: Option<(FieldId, FieldId)>,
 ) -> Result<ExtractedFacetValues> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();

    let mut fid_docid_facet_numbers_sorter = create_sorter(
@ -127,12 +124,18 @@ pub fn extract_fid_docid_facet_values<R: io::Read + io::Seek>(
                    add_exists.insert(document);
                }

-                let geo_support =
-                    geo_fields_ids.map_or(false, |(lat, lng)| field_id == lat || field_id == lng);
+                let del_geo_support = settings_diff
+                    .old
+                    .geo_fields_ids
+                    .map_or(false, |(lat, lng)| field_id == lat || field_id == lng);
+                let add_geo_support = settings_diff
+                    .new
+                    .geo_fields_ids
+                    .map_or(false, |(lat, lng)| field_id == lat || field_id == lng);
                let del_filterable_values =
-                    del_value.map(|value| extract_facet_values(&value, geo_support));
+                    del_value.map(|value| extract_facet_values(&value, del_geo_support));
                let add_filterable_values =
-                    add_value.map(|value| extract_facet_values(&value, geo_support));
+                    add_value.map(|value| extract_facet_values(&value, add_geo_support));

                // Those closures are just here to simplify things a bit.
                let mut insert_numbers_diff = |del_numbers, add_numbers| {
--- a/milli/src/update/index_documents/extract/extract_fid_word_count_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_fid_word_count_docids.rs
@ -26,8 +26,6 @@ pub fn extract_fid_word_count_docids<R: io::Read + io::Seek>(
    indexer: GrenadParameters,
    _settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();

    let mut fid_word_count_docids_sorter = create_sorter(
--- a/milli/src/update/index_documents/extract/extract_geo_points.rs
+++ b/milli/src/update/index_documents/extract/extract_geo_points.rs
@ -8,6 +8,7 @@ use super::helpers::{create_writer, writer_into_reader, GrenadParameters};
 use crate::error::GeoError;
 use crate::update::del_add::{DelAdd, KvReaderDelAdd, KvWriterDelAdd};
 use crate::update::index_documents::extract_finite_float_from_value;
+use crate::update::settings::{InnerIndexSettings, InnerIndexSettingsDiff};
 use crate::{FieldId, InternalError, Result};

 /// Extracts the geographical coordinates contained in each document under the `_geo` field.
@ -18,10 +19,8 @@ pub fn extract_geo_points<R: io::Read + io::Seek>(
    obkv_documents: grenad::Reader<R>,
    indexer: GrenadParameters,
    primary_key_id: FieldId,
-    (lat_fid, lng_fid): (FieldId, FieldId),
+    settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
-
    let mut writer = create_writer(
        indexer.chunk_compression_type,
        indexer.chunk_compression_level,
@ -40,47 +39,27 @@ pub fn extract_geo_points<R: io::Read + io::Seek>(
            serde_json::from_slice(document_id).unwrap()
        };

-        // first we get the two fields
-        match (obkv.get(lat_fid), obkv.get(lng_fid)) {
-            (Some(lat), Some(lng)) => {
-                let deladd_lat_obkv = KvReaderDelAdd::new(lat);
-                let deladd_lng_obkv = KvReaderDelAdd::new(lng);
+        // extract old version
+        let del_lat_lng =
+            extract_lat_lng(&obkv, &settings_diff.old, DelAdd::Deletion, document_id)?;
+        // extract new version
+        let add_lat_lng =
+            extract_lat_lng(&obkv, &settings_diff.new, DelAdd::Addition, document_id)?;

-                // then we extract the values
-                let del_lat_lng = deladd_lat_obkv
-                    .get(DelAdd::Deletion)
-                    .zip(deladd_lng_obkv.get(DelAdd::Deletion))
-                    .map(|(lat, lng)| extract_lat_lng(lat, lng, document_id))
-                    .transpose()?;
-                let add_lat_lng = deladd_lat_obkv
-                    .get(DelAdd::Addition)
-                    .zip(deladd_lng_obkv.get(DelAdd::Addition))
-                    .map(|(lat, lng)| extract_lat_lng(lat, lng, document_id))
-                    .transpose()?;
-
-                if del_lat_lng != add_lat_lng {
-                    let mut obkv = KvWriterDelAdd::memory();
-                    if let Some([lat, lng]) = del_lat_lng {
-                        #[allow(clippy::drop_non_drop)]
-                        let bytes: [u8; 16] = concat_arrays![lat.to_ne_bytes(), lng.to_ne_bytes()];
-                        obkv.insert(DelAdd::Deletion, bytes)?;
-                    }
-                    if let Some([lat, lng]) = add_lat_lng {
-                        #[allow(clippy::drop_non_drop)]
-                        let bytes: [u8; 16] = concat_arrays![lat.to_ne_bytes(), lng.to_ne_bytes()];
-                        obkv.insert(DelAdd::Addition, bytes)?;
-                    }
-                    let bytes = obkv.into_inner()?;
-                    writer.insert(docid_bytes, bytes)?;
-                }
+        if del_lat_lng != add_lat_lng {
+            let mut obkv = KvWriterDelAdd::memory();
+            if let Some([lat, lng]) = del_lat_lng {
+                #[allow(clippy::drop_non_drop)]
+                let bytes: [u8; 16] = concat_arrays![lat.to_ne_bytes(), lng.to_ne_bytes()];
+                obkv.insert(DelAdd::Deletion, bytes)?;
            }
-            (None, Some(_)) => {
-                return Err(GeoError::MissingLatitude { document_id: document_id() }.into())
+            if let Some([lat, lng]) = add_lat_lng {
+                #[allow(clippy::drop_non_drop)]
+                let bytes: [u8; 16] = concat_arrays![lat.to_ne_bytes(), lng.to_ne_bytes()];
+                obkv.insert(DelAdd::Addition, bytes)?;
            }
-            (Some(_), None) => {
-                return Err(GeoError::MissingLongitude { document_id: document_id() }.into())
-            }
-            (None, None) => (),
+            let bytes = obkv.into_inner()?;
+            writer.insert(docid_bytes, bytes)?;
        }
    }

@ -88,16 +67,37 @@ pub fn extract_geo_points<R: io::Read + io::Seek>(
 }

 /// Extract the finite floats lat and lng from two bytes slices.
-fn extract_lat_lng(lat: &[u8], lng: &[u8], document_id: impl Fn() -> Value) -> Result<[f64; 2]> {
-    let lat = extract_finite_float_from_value(
-        serde_json::from_slice(lat).map_err(InternalError::SerdeJson)?,
-    )
-    .map_err(|lat| GeoError::BadLatitude { document_id: document_id(), value: lat })?;
+fn extract_lat_lng(
+    document: &obkv::KvReader<FieldId>,
+    settings: &InnerIndexSettings,
+    deladd: DelAdd,
+    document_id: impl Fn() -> Value,
+) -> Result<Option<[f64; 2]>> {
+    match settings.geo_fields_ids {
+        Some((lat_fid, lng_fid)) => {
+            let lat = document.get(lat_fid).map(KvReaderDelAdd::new).and_then(|r| r.get(deladd));
+            let lng = document.get(lng_fid).map(KvReaderDelAdd::new).and_then(|r| r.get(deladd));
+            let (lat, lng) = match (lat, lng) {
+                (Some(lat), Some(lng)) => (lat, lng),
+                (Some(_), None) => {
+                    return Err(GeoError::MissingLatitude { document_id: document_id() }.into())
+                }
+                (None, Some(_)) => {
+                    return Err(GeoError::MissingLongitude { document_id: document_id() }.into())
+                }
+                (None, None) => return Ok(None),
+            };
+            let lat = extract_finite_float_from_value(
+                serde_json::from_slice(lat).map_err(InternalError::SerdeJson)?,
+            )
+            .map_err(|lat| GeoError::BadLatitude { document_id: document_id(), value: lat })?;

-    let lng = extract_finite_float_from_value(
-        serde_json::from_slice(lng).map_err(InternalError::SerdeJson)?,
-    )
-    .map_err(|lng| GeoError::BadLongitude { document_id: document_id(), value: lng })?;
-
-    Ok([lat, lng])
+            let lng = extract_finite_float_from_value(
+                serde_json::from_slice(lng).map_err(InternalError::SerdeJson)?,
+            )
+            .map_err(|lng| GeoError::BadLongitude { document_id: document_id(), value: lng })?;
+            Ok(Some([lat, lng]))
+        }
+        None => Ok(None),
+    }
 }
--- a/milli/src/update/index_documents/extract/extract_vector_points.rs
+++ b/milli/src/update/index_documents/extract/extract_vector_points.rs
@ -10,16 +10,16 @@ use bytemuck::cast_slice;
 use grenad::Writer;
 use itertools::EitherOrBoth;
 use ordered_float::OrderedFloat;
-use serde_json::{from_slice, Value};
+use serde_json::Value;

 use super::helpers::{create_writer, writer_into_reader, GrenadParameters};
-use crate::error::UserError;
 use crate::prompt::Prompt;
 use crate::update::del_add::{DelAdd, KvReaderDelAdd, KvWriterDelAdd};
 use crate::update::index_documents::helpers::try_split_at;
 use crate::update::settings::InnerIndexSettingsDiff;
+use crate::vector::parsed_vectors::{ParsedVectorsDiff, RESERVED_VECTORS_FIELD_NAME};
 use crate::vector::Embedder;
-use crate::{DocumentId, InternalError, Result, ThreadPoolNoAbort, VectorOrArrayOfVectors};
+use crate::{DocumentId, Result, ThreadPoolNoAbort};

 /// The length of the elements that are always in the buffer when inserting new values.
 const TRUNCATE_SIZE: usize = size_of::<DocumentId>();
@ -31,6 +31,10 @@ pub struct ExtractedVectorPoints {
    pub remove_vectors: grenad::Reader<BufReader<File>>,
    // docid -> prompt
    pub prompts: grenad::Reader<BufReader<File>>,
+
+    // embedder
+    pub embedder_name: String,
+    pub embedder: Arc<Embedder>,
 }

 enum VectorStateDelta {
@ -65,6 +69,19 @@ impl VectorStateDelta {
    }
 }

+struct EmbedderVectorExtractor {
+    embedder_name: String,
+    embedder: Arc<Embedder>,
+    prompt: Arc<Prompt>,
+
+    // (docid, _index) -> KvWriterDelAdd -> Vector
+    manual_vectors_writer: Writer<BufWriter<File>>,
+    // (docid) -> (prompt)
+    prompts_writer: Writer<BufWriter<File>>,
+    // (docid) -> ()
+    remove_vectors_writer: Writer<BufWriter<File>>,
+}
+
 /// Extracts the embedding vector contained in each document under the `_vectors` field.
 ///
 /// Returns the generated grenad reader containing the docid as key associated to the Vec<f32>
@ -73,34 +90,52 @@ pub fn extract_vector_points<R: io::Read + io::Seek>(
    obkv_documents: grenad::Reader<R>,
    indexer: GrenadParameters,
    settings_diff: &InnerIndexSettingsDiff,
-    prompt: &Prompt,
-    embedder_name: &str,
-) -> Result<ExtractedVectorPoints> {
-    puffin::profile_function!();
+) -> Result<Vec<ExtractedVectorPoints>> {
+    let reindex_vectors = settings_diff.reindex_vectors();

    let old_fields_ids_map = &settings_diff.old.fields_ids_map;
    let new_fields_ids_map = &settings_diff.new.fields_ids_map;
+    // the vector field id may have changed
+    let old_vectors_fid = old_fields_ids_map.id(RESERVED_VECTORS_FIELD_NAME);
+    // filter the old vector fid if the settings has been changed forcing reindexing.
+    let old_vectors_fid = old_vectors_fid.filter(|_| !reindex_vectors);

-    // (docid, _index) -> KvWriterDelAdd -> Vector
-    let mut manual_vectors_writer = create_writer(
-        indexer.chunk_compression_type,
-        indexer.chunk_compression_level,
-        tempfile::tempfile()?,
-    );
+    let new_vectors_fid = new_fields_ids_map.id(RESERVED_VECTORS_FIELD_NAME);

-    // (docid) -> (prompt)
-    let mut prompts_writer = create_writer(
-        indexer.chunk_compression_type,
-        indexer.chunk_compression_level,
-        tempfile::tempfile()?,
-    );
+    let mut extractors = Vec::new();
+    for (embedder_name, (embedder, prompt)) in
+        settings_diff.new.embedding_configs.clone().into_iter()
+    {
+        // (docid, _index) -> KvWriterDelAdd -> Vector
+        let manual_vectors_writer = create_writer(
+            indexer.chunk_compression_type,
+            indexer.chunk_compression_level,
+            tempfile::tempfile()?,
+        );

-    // (docid) -> ()
-    let mut remove_vectors_writer = create_writer(
-        indexer.chunk_compression_type,
-        indexer.chunk_compression_level,
-        tempfile::tempfile()?,
-    );
+        // (docid) -> (prompt)
+        let prompts_writer = create_writer(
+            indexer.chunk_compression_type,
+            indexer.chunk_compression_level,
+            tempfile::tempfile()?,
+        );
+
+        // (docid) -> ()
+        let remove_vectors_writer = create_writer(
+            indexer.chunk_compression_type,
+            indexer.chunk_compression_level,
+            tempfile::tempfile()?,
+        );
+
+        extractors.push(EmbedderVectorExtractor {
+            embedder_name,
+            embedder,
+            prompt,
+            manual_vectors_writer,
+            prompts_writer,
+            remove_vectors_writer,
+        });
+    }

    let mut key_buffer = Vec::new();
    let mut cursor = obkv_documents.into_cursor()?;
@ -114,152 +149,138 @@ pub fn extract_vector_points<R: io::Read + io::Seek>(
        key_buffer.clear();
        key_buffer.extend_from_slice(docid_bytes);

-        // since we only needs the primary key when we throw an error we create this getter to
+        // since we only need the primary key when we throw an error we create this getter to
        // lazily get it when needed
        let document_id = || -> Value { from_utf8(external_id_bytes).unwrap().into() };

-        // the vector field id may have changed
-        let old_vectors_fid = old_fields_ids_map.id("_vectors");
-        // filter the old vector fid if the settings has been changed forcing reindexing.
-        let old_vectors_fid = old_vectors_fid.filter(|_| !settings_diff.reindex_vectors());
+        let mut parsed_vectors = ParsedVectorsDiff::new(obkv, old_vectors_fid, new_vectors_fid)
+            .map_err(|error| error.to_crate_error(document_id().to_string()))?;

-        let new_vectors_fid = new_fields_ids_map.id("_vectors");
-        let vectors_field = {
-            let del = old_vectors_fid
-                .and_then(|vectors_fid| obkv.get(vectors_fid))
-                .map(KvReaderDelAdd::new)
-                .map(|obkv| to_vector_map(obkv, DelAdd::Deletion, &document_id))
-                .transpose()?
-                .flatten();
-            let add = new_vectors_fid
-                .and_then(|vectors_fid| obkv.get(vectors_fid))
-                .map(KvReaderDelAdd::new)
-                .map(|obkv| to_vector_map(obkv, DelAdd::Addition, &document_id))
-                .transpose()?
-                .flatten();
-            (del, add)
-        };
+        for EmbedderVectorExtractor {
+            embedder_name,
+            embedder: _,
+            prompt,
+            manual_vectors_writer,
+            prompts_writer,
+            remove_vectors_writer,
+        } in extractors.iter_mut()
+        {
+            let delta = match parsed_vectors.remove(embedder_name) {
+                (Some(old), Some(new)) => {
+                    // no autogeneration
+                    let del_vectors = old.into_array_of_vectors();
+                    let add_vectors = new.into_array_of_vectors();

-        let (del_map, add_map) = vectors_field;
-
-        let del_value = del_map.and_then(|mut map| map.remove(embedder_name));
-        let add_value = add_map.and_then(|mut map| map.remove(embedder_name));
-
-        let delta = match (del_value, add_value) {
-            (Some(old), Some(new)) => {
-                // no autogeneration
-                let del_vectors = extract_vectors(old, document_id, embedder_name)?;
-                let add_vectors = extract_vectors(new, document_id, embedder_name)?;
-
-                if add_vectors.len() > usize::from(u8::MAX) {
-                    return Err(crate::Error::UserError(crate::UserError::TooManyVectors(
-                        document_id().to_string(),
-                        add_vectors.len(),
-                    )));
-                }
-
-                VectorStateDelta::ManualDelta(del_vectors, add_vectors)
-            }
-            (Some(_old), None) => {
-                // Do we keep this document?
-                let document_is_kept = obkv
-                    .iter()
-                    .map(|(_, deladd)| KvReaderDelAdd::new(deladd))
-                    .any(|deladd| deladd.get(DelAdd::Addition).is_some());
-                if document_is_kept {
-                    // becomes autogenerated
-                    VectorStateDelta::NowGenerated(prompt.render(
-                        obkv,
-                        DelAdd::Addition,
-                        new_fields_ids_map,
-                    )?)
-                } else {
-                    VectorStateDelta::NowRemoved
-                }
-            }
-            (None, Some(new)) => {
-                // was possibly autogenerated, remove all vectors for that document
-                let add_vectors = extract_vectors(new, document_id, embedder_name)?;
-                if add_vectors.len() > usize::from(u8::MAX) {
-                    return Err(crate::Error::UserError(crate::UserError::TooManyVectors(
-                        document_id().to_string(),
-                        add_vectors.len(),
-                    )));
-                }
-
-                VectorStateDelta::WasGeneratedNowManual(add_vectors)
-            }
-            (None, None) => {
-                // Do we keep this document?
-                let document_is_kept = obkv
-                    .iter()
-                    .map(|(_, deladd)| KvReaderDelAdd::new(deladd))
-                    .any(|deladd| deladd.get(DelAdd::Addition).is_some());
-
-                if document_is_kept {
-                    // Don't give up if the old prompt was failing
-                    let old_prompt = Some(prompt)
-                        // TODO: this filter works because we erase the vec database when a embedding setting changes.
-                        // When vector pipeline will be optimized, this should be removed.
-                        .filter(|_| !settings_diff.reindex_vectors())
-                        .map(|p| {
-                            p.render(obkv, DelAdd::Deletion, old_fields_ids_map).unwrap_or_default()
-                        });
-                    let new_prompt = prompt.render(obkv, DelAdd::Addition, new_fields_ids_map)?;
-                    if old_prompt.as_ref() != Some(&new_prompt) {
-                        let old_prompt = old_prompt.unwrap_or_default();
-                        tracing::trace!(
-                            "🚀 Changing prompt from\n{old_prompt}\n===to===\n{new_prompt}"
-                        );
-                        VectorStateDelta::NowGenerated(new_prompt)
-                    } else {
-                        tracing::trace!("⏭️ Prompt unmodified, skipping");
-                        VectorStateDelta::NoChange
+                    if add_vectors.len() > usize::from(u8::MAX) {
+                        return Err(crate::Error::UserError(crate::UserError::TooManyVectors(
+                            document_id().to_string(),
+                            add_vectors.len(),
+                        )));
                    }
-                } else {
-                    VectorStateDelta::NowRemoved
-                }
-            }
-        };

-        // and we finally push the unique vectors into the writer
-        push_vectors_diff(
-            &mut remove_vectors_writer,
-            &mut prompts_writer,
-            &mut manual_vectors_writer,
-            &mut key_buffer,
-            delta,
-            settings_diff,
-        )?;
+                    VectorStateDelta::ManualDelta(del_vectors, add_vectors)
+                }
+                (Some(_old), None) => {
+                    // Do we keep this document?
+                    let document_is_kept = obkv
+                        .iter()
+                        .map(|(_, deladd)| KvReaderDelAdd::new(deladd))
+                        .any(|deladd| deladd.get(DelAdd::Addition).is_some());
+                    if document_is_kept {
+                        // becomes autogenerated
+                        VectorStateDelta::NowGenerated(prompt.render(
+                            obkv,
+                            DelAdd::Addition,
+                            new_fields_ids_map,
+                        )?)
+                    } else {
+                        VectorStateDelta::NowRemoved
+                    }
+                }
+                (None, Some(new)) => {
+                    // was possibly autogenerated, remove all vectors for that document
+                    let add_vectors = new.into_array_of_vectors();
+                    if add_vectors.len() > usize::from(u8::MAX) {
+                        return Err(crate::Error::UserError(crate::UserError::TooManyVectors(
+                            document_id().to_string(),
+                            add_vectors.len(),
+                        )));
+                    }
+
+                    VectorStateDelta::WasGeneratedNowManual(add_vectors)
+                }
+                (None, None) => {
+                    // Do we keep this document?
+                    let document_is_kept = obkv
+                        .iter()
+                        .map(|(_, deladd)| KvReaderDelAdd::new(deladd))
+                        .any(|deladd| deladd.get(DelAdd::Addition).is_some());
+
+                    if document_is_kept {
+                        // Don't give up if the old prompt was failing
+                        let old_prompt = Some(&prompt)
+                            // TODO: this filter works because we erase the vec database when a embedding setting changes.
+                            // When vector pipeline will be optimized, this should be removed.
+                            .filter(|_| !settings_diff.reindex_vectors())
+                            .map(|p| {
+                                p.render(obkv, DelAdd::Deletion, old_fields_ids_map)
+                                    .unwrap_or_default()
+                            });
+                        let new_prompt =
+                            prompt.render(obkv, DelAdd::Addition, new_fields_ids_map)?;
+                        if old_prompt.as_ref() != Some(&new_prompt) {
+                            let old_prompt = old_prompt.unwrap_or_default();
+                            tracing::trace!(
+                                "🚀 Changing prompt from\n{old_prompt}\n===to===\n{new_prompt}"
+                            );
+                            VectorStateDelta::NowGenerated(new_prompt)
+                        } else {
+                            tracing::trace!("⏭️ Prompt unmodified, skipping");
+                            VectorStateDelta::NoChange
+                        }
+                    } else {
+                        VectorStateDelta::NowRemoved
+                    }
+                }
+            };
+
+            // and we finally push the unique vectors into the writer
+            push_vectors_diff(
+                remove_vectors_writer,
+                prompts_writer,
+                manual_vectors_writer,
+                &mut key_buffer,
+                delta,
+                reindex_vectors,
+            )?;
+        }
    }

-    Ok(ExtractedVectorPoints {
-        // docid, _index -> KvWriterDelAdd -> Vector
-        manual_vectors: writer_into_reader(manual_vectors_writer)?,
-        // docid -> ()
-        remove_vectors: writer_into_reader(remove_vectors_writer)?,
-        // docid -> prompt
-        prompts: writer_into_reader(prompts_writer)?,
-    })
-}
+    let mut results = Vec::new();

-fn to_vector_map(
-    obkv: KvReaderDelAdd,
-    side: DelAdd,
-    document_id: &impl Fn() -> Value,
-) -> Result<Option<serde_json::Map<String, Value>>> {
-    Ok(if let Some(value) = obkv.get(side) {
-        let Ok(value) = from_slice(value) else {
-            let value = from_slice(value).map_err(InternalError::SerdeJson)?;
-            return Err(crate::Error::UserError(UserError::InvalidVectorsMapType {
-                document_id: document_id(),
-                value,
-            }));
-        };
-        Some(value)
-    } else {
-        None
-    })
+    for EmbedderVectorExtractor {
+        embedder_name,
+        embedder,
+        prompt: _,
+        manual_vectors_writer,
+        prompts_writer,
+        remove_vectors_writer,
+    } in extractors
+    {
+        results.push(ExtractedVectorPoints {
+            // docid, _index -> KvWriterDelAdd -> Vector
+            manual_vectors: writer_into_reader(manual_vectors_writer)?,
+            // docid -> ()
+            remove_vectors: writer_into_reader(remove_vectors_writer)?,
+            // docid -> prompt
+            prompts: writer_into_reader(prompts_writer)?,
+
+            embedder,
+            embedder_name,
+        })
+    }
+
+    Ok(results)
 }

 /// Computes the diff between both Del and Add numbers and
@ -270,14 +291,13 @@ fn push_vectors_diff(
    manual_vectors_writer: &mut Writer<BufWriter<File>>,
    key_buffer: &mut Vec<u8>,
    delta: VectorStateDelta,
-    settings_diff: &InnerIndexSettingsDiff,
+    reindex_vectors: bool,
 ) -> Result<()> {
-    puffin::profile_function!();
    let (must_remove, prompt, (mut del_vectors, mut add_vectors)) = delta.into_values();
    if must_remove
    // TODO: the below condition works because we erase the vec database when a embedding setting changes.
    // When vector pipeline will be optimized, this should be removed.
-    && !settings_diff.reindex_vectors()
+    && !reindex_vectors
    {
        key_buffer.truncate(TRUNCATE_SIZE);
        remove_vectors_writer.insert(&key_buffer, [])?;
@ -308,7 +328,7 @@ fn push_vectors_diff(
            EitherOrBoth::Left(vector) => {
                // TODO: the below condition works because we erase the vec database when a embedding setting changes.
                // When vector pipeline will be optimized, this should be removed.
-                if !settings_diff.reindex_vectors() {
+                if !reindex_vectors {
                    // We insert only the Del part of the Obkv to inform
                    // that we only want to remove all those vectors.
                    let mut obkv = KvWriterDelAdd::memory();
@ -336,26 +356,6 @@ fn compare_vectors(a: &[f32], b: &[f32]) -> Ordering {
    a.iter().copied().map(OrderedFloat).cmp(b.iter().copied().map(OrderedFloat))
 }

-/// Extracts the vectors from a JSON value.
-fn extract_vectors(
-    value: Value,
-    document_id: impl Fn() -> Value,
-    name: &str,
-) -> Result<Vec<Vec<f32>>> {
-    // FIXME: ugly clone of the vectors here
-    match serde_json::from_value(value.clone()) {
-        Ok(vectors) => {
-            Ok(VectorOrArrayOfVectors::into_array_of_vectors(vectors).unwrap_or_default())
-        }
-        Err(_) => Err(UserError::InvalidVectorsType {
-            document_id: document_id(),
-            value,
-            subfield: name.to_owned(),
-        }
-        .into()),
-    }
-}
-
 #[tracing::instrument(level = "trace", skip_all, target = "indexing::extract")]
 pub fn extract_embeddings<R: io::Read + io::Seek>(
    // docid, prompt
@ -364,7 +364,6 @@ pub fn extract_embeddings<R: io::Read + io::Seek>(
    embedder: Arc<Embedder>,
    request_threads: &ThreadPoolNoAbort,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
    let n_chunks = embedder.chunk_count_hint(); // chunk level parallelism
    let n_vectors_per_chunk = embedder.prompt_count_in_chunk_hint(); // number of vectors in a single chunk

--- a/milli/src/update/index_documents/extract/extract_word_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_word_docids.rs
@ -36,8 +36,6 @@ pub fn extract_word_docids<R: io::Read + io::Seek>(
    grenad::Reader<BufReader<File>>,
    grenad::Reader<BufReader<File>>,
 )> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();

    let mut word_fid_docids_sorter = create_sorter(
@ -167,8 +165,6 @@ fn words_into_sorter(
    add_words: &BTreeSet<Vec<u8>>,
    word_fid_docids_sorter: &mut grenad::Sorter<MergeFn>,
 ) -> Result<()> {
-    puffin::profile_function!();
-
    use itertools::merge_join_by;
    use itertools::EitherOrBoth::{Both, Left, Right};

--- a/milli/src/update/index_documents/extract/extract_word_pair_proximity_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_word_pair_proximity_docids.rs
@ -26,7 +26,6 @@ pub fn extract_word_pair_proximity_docids<R: io::Read + io::Seek>(
    indexer: GrenadParameters,
    settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
    let any_deletion = settings_diff.old.proximity_precision == ProximityPrecision::ByWord;
    let any_addition = settings_diff.new.proximity_precision == ProximityPrecision::ByWord;

@ -71,8 +70,6 @@ pub fn extract_word_pair_proximity_docids<R: io::Read + io::Seek>(

        // if we change document, we fill the sorter
        if current_document_id.map_or(false, |id| id != document_id) {
-            puffin::profile_scope!("Document into sorter");
-
            // FIXME: span inside of a hot loop might degrade performance and create big reports
            let span = tracing::trace_span!(target: "indexing::details", "document_into_sorter");
            let _entered = span.enter();
@ -163,7 +160,6 @@ pub fn extract_word_pair_proximity_docids<R: io::Read + io::Seek>(
    }

    if let Some(document_id) = current_document_id {
-        puffin::profile_scope!("Final document into sorter");
        // FIXME: span inside of a hot loop might degrade performance and create big reports
        let span = tracing::trace_span!(target: "indexing::details", "final_document_into_sorter");
        let _entered = span.enter();
@ -176,7 +172,6 @@ pub fn extract_word_pair_proximity_docids<R: io::Read + io::Seek>(
        )?;
    }
    {
-        puffin::profile_scope!("sorter_into_reader");
        // FIXME: span inside of a hot loop might degrade performance and create big reports
        let span = tracing::trace_span!(target: "indexing::details", "sorter_into_reader");
        let _entered = span.enter();
--- a/milli/src/update/index_documents/extract/extract_word_position_docids.rs
+++ b/milli/src/update/index_documents/extract/extract_word_position_docids.rs
@ -25,8 +25,6 @@ pub fn extract_word_position_docids<R: io::Read + io::Seek>(
    indexer: GrenadParameters,
    _settings_diff: &InnerIndexSettingsDiff,
 ) -> Result<grenad::Reader<BufReader<File>>> {
-    puffin::profile_function!();
-
    let max_memory = indexer.max_memory_by_thread();

    let mut word_position_docids_sorter = create_sorter(
@ -104,8 +102,6 @@ fn words_position_into_sorter(
    add_word_positions: &BTreeSet<(u16, Vec<u8>)>,
    word_position_docids_sorter: &mut grenad::Sorter<MergeFn>,
 ) -> Result<()> {
-    puffin::profile_function!();
-
    use itertools::merge_join_by;
    use itertools::EitherOrBoth::{Both, Left, Right};

--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
Clément Renault	af5b1a88a4	Expose a route to show advanced index stats	2024-06-04 14:08:01 -04:00
meili-bors[bot]	d6bd88ce4f	Merge #4667 4667: Frequency matching strategy r=Kerollmops a=ManyTheFish # Pull Request ## Related issue Fixes #3773 ## What does this PR do? - add test for matching strategy - implement frequency matching strategy See the [PRD for more details](https://www.notion.so/meilisearch/Frequency-Matching-Strategy-0f3ba08833a442a39590a53a1505ab00). [Public API](https://www.notion.so/meilisearch/frequency-matching-strategy-89868fb7fc584026bc56e378eb854a7f). Co-authored-by: ManyTheFish <many@meilisearch.com>	2024-05-30 14:53:31 +00:00
ManyTheFish	3f1a510069	Add tests and fix matching strategy	2024-05-30 12:02:42 +02:00
meili-bors[bot]	06a9803544	Merge #4664 4664: Update README.md r=curquiza a=tpayet Add hybrid & semantic as a feature # Pull Request ## Related issue Fixes #<issue_number> ## What does this PR do? - ... ## PR checklist Please check if your PR fulfills the following requirements: - [ ] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)? - [ ] Have you read the contributing guidelines? - [ ] Have you made sure that the title is accurate and descriptive of the changes? Thank you so much for contributing to Meilisearch! Co-authored-by: Thomas Payet <thomas@meilisearch.com>	2024-05-29 16:55:20 +00:00
Thomas Payet	b2588d8101	Update README.md Add hybrid & semantic as a feature	2024-05-29 17:48:48 +02:00
meili-bors[bot]	62d27172f4	Merge #4663 4663: Bring back release v1.8.1 into main r=ManyTheFish a=ManyTheFish Co-authored-by: Tamo <tamo@meilisearch.com> Co-authored-by: ManyTheFish <many@meilisearch.com> Co-authored-by: meili-bors[bot] <89034592+meili-bors[bot]@users.noreply.github.com> Co-authored-by: ManyTheFish <ManyTheFish@users.noreply.github.com> Co-authored-by: Many the fish <many@meilisearch.com>	2024-05-29 14:47:38 +00:00
ManyTheFish	1ab88e10b9	Merge branch 'main' into merge-release-v1.8.1-in-main	2024-05-29 16:24:00 +02:00
ManyTheFish	6a4b2516aa	WIP	2024-05-29 16:21:24 +02:00
ManyTheFish	abdc4afcca	Implement Frequency matching strategy	2024-05-29 13:59:08 +02:00
meili-bors[bot]	75d5c0ae1f	Merge #4647 4647: Feature: get similar documents r=dureuill a=dureuill # Pull Request ## Related issue Fixes #4610 ## What does this PR do? [Usage](https://meilisearch.notion.site/Get-similar-documents-usage-540919ca755c4da0b7cdee273db3f290) Co-authored-by: Louis Dureuil <louis@meilisearch.com>	2024-05-29 11:42:23 +00:00
meili-bors[bot]	a88554216a	Merge #4657 4657: Update version for the next release (v1.9.0) in Cargo.toml r=curquiza a=meili-bot ⚠️ This PR is automatically generated. Check the new version is the expected one and Cargo.lock has been updated before merging. Co-authored-by: curquiza <curquiza@users.noreply.github.com>	2024-05-29 11:14:19 +00:00
Louis Dureuil	2cf3e1c80a	Temporarily ignore perform snapshot test under Windows	2024-05-29 12:42:47 +02:00
Many the fish	e1fbfde6c4	Merge branch 'main' into merge-release-v1.8.1-in-main	2024-05-29 11:31:03 +02:00
ManyTheFish	27b75ec648	merge main into v1.8.1	2024-05-29 11:26:07 +02:00
curquiza	07fdb081a4	Update version for the next release (v1.9.0) in Cargo.toml	2024-05-28 14:19:40 +00:00
Louis Dureuil	ca006e38ec	Basic tests	2024-05-28 15:28:19 +02:00
Louis Dureuil	e26bd87780	Error tests for similar routes	2024-05-28 15:28:19 +02:00
Louis Dureuil	c01e498a63	Test server can call similar	2024-05-28 15:28:19 +02:00
Louis Dureuil	ca6cc4654b	Add similar route	2024-05-28 15:28:19 +02:00
Louis Dureuil	3bd9d2478c	Add error codes	2024-05-28 15:27:43 +02:00
Louis Dureuil	54b15059a0	Analytics changes	2024-05-28 15:27:43 +02:00
Louis Dureuil	d35278320e	Add support functions for accessing arroy writers and readers	2024-05-28 15:27:43 +02:00
Louis Dureuil	e172e938e7	add search rules directly takes the filter rather than the searchquery	2024-05-28 15:22:25 +02:00
Louis Dureuil	02b3d82c60	filtered_universe accepts index and txn instead of SearchContext	2024-05-28 15:22:12 +02:00
Louis Dureuil	fd2c95999d	Change `validate_document_id` to public and remove extra layer of result	2024-05-28 15:21:19 +02:00
meili-bors[bot]	e248d2a1e6	Merge #4655 4655: Remove `exportPuffinReport` experimental feature r=Kerollmops a=Kerollmops This PR fixes #4605 by removing every trace of Puffin. Puffin is a great tool, but we use a better approach to measuring performance. Co-authored-by: Clément Renault <clement@meilisearch.com>	2024-05-28 07:01:16 +00:00
Clément Renault	487431a035	Fix tests	2024-05-27 16:12:20 +02:00
Clément Renault	b6d450d484	Remove puffin experimental feature	2024-05-27 15:59:28 +02:00
Clément Renault	dc949ab46a	Remove puffin usage	2024-05-27 15:59:14 +02:00
Clément Renault	7f3e51349e	Remove puffin for the dependencies	2024-05-27 15:53:06 +02:00
meili-bors[bot]	19acc65ad2	Merge #4646 4646: Reduce `Transform`'s disk usage r=Kerollmops a=Kerollmops This PR implements what is described in #4485. It reduces the number of disk writes and disk usage. Co-authored-by: Clément Renault <clement@meilisearch.com>	2024-05-23 16:06:50 +00:00
meili-bors[bot]	3a3ab17714	Merge #4651 4651: Allow to comment with the results of benchmark invocation r=Kerollmops a=dureuill Co-authored-by: Louis Dureuil <louis@meilisearch.com>	2024-05-23 15:32:09 +00:00
Louis Dureuil	eaf57056ca	comment with the results of benchmarks	2024-05-23 15:34:39 +02:00
Louis Dureuil	e340705634	Change benchmark outputs - logs to stderr instead of stdout - prints links to the dashboard when there is a dashboard	2024-05-23 15:29:06 +02:00
Clément Renault	fe17c0f52e	Construct the minimal OBKVs according to the settings diff	2024-05-23 11:23:57 +02:00
meili-bors[bot]	14bc80e3df	Merge #4633 4633: Allow to mark vectors as "userProvided" r=Kerollmops a=dureuill # Pull Request ## Related issue Fixes #4606 ## What does this PR do? [See usage in PRD](https://meilisearch.notion.site/v1-9-AI-search-changes-e90d6803eca8417aa70a1ac5d0225697#deb96fb0595947bda7d4a371100326eb) - Extends the shape of the special `_vectors` field in documents. - previously, the `_vectors` field had to be an object, with each field the name of a configured embedder, and each value either `null`, an embedding (array of numbers), or an array of embeddings. - In this PR, the value of an embedder in the `_vectors` field can additionally be an object. The object has two fields: 1. `embeddings`: `null`, an embedding (array of numbers), or an array of embeddings. 2. `userProvided`: a boolean indicating if the vector was provided by the user. - The previous form `embedder_or_array_of_embedders` is semantically equivalent to: ```json { "embeddings": embedder_or_array_of_embedders, "userProvided": true } ``` - During the indexing step, the subfields and values of the `_vectors` field that have `userProvided` set to false are added in the vector DB, but not in the documents DB: that means that future modifications of the documents will trigger a regeneration of that particular vector using the document template. - This allows importing embeddings as a one-shot process, while still retaining the ability to regenerate embeddings on document change. - The dump process now uses this ability: it enriches the `_vectors` fields of documents with the embeddings that were autogenerated, marking them as not `userProvided`. This allows importing the vectors from a dump without regenerating them. ### Tests This PR adds the following tests - Long-needed hybrid search tests of a simple hf embedder - Dump test that imports vectors. Due to the difficulty of actually importing a dump in tests, we just read the dump and check it contains the expected content. - Tests in the index-scheduler: this tests that documents containing the same kind of instructions as in the dump indexes as expected Co-authored-by: Louis Dureuil <louis@meilisearch.com>	2024-05-23 08:17:54 +00:00
Clément Renault	bc5663e673	FieldIdsMap no longer useful thanks to #4631	2024-05-22 16:06:15 +02:00
Louis Dureuil	8a941c0241	Smaller review changes	2024-05-22 14:44:42 +02:00
Louis Dureuil	3412e7fbcf	"[]" is deserialized as 0 embedding rather than 1 embedding of dim 0	2024-05-22 12:25:21 +02:00
Louis Dureuil	16037e2169	Don't remove embedders that are not in the config from the document DB	2024-05-22 12:24:51 +02:00
Louis Dureuil	8f7c8ca7f0	Remove now unused error variant	2024-05-22 12:23:43 +02:00
meili-bors[bot]	ba75d23bfe	Merge #4648 4648: Update version for the next release (v1.8.1) in Cargo.toml r=ManyTheFish a=meili-bot ⚠️ This PR is automatically generated. Check the new version is the expected one and Cargo.lock has been updated before merging. Co-authored-by: ManyTheFish <ManyTheFish@users.noreply.github.com>	2024-05-21 16:38:36 +00:00
ManyTheFish	7fbb3bf8e8	Update version for the next release (v1.8.1) in Cargo.toml	2024-05-21 15:13:03 +00:00
Clément Renault	500ddc76b5	Make the flattened sorter optional	2024-05-21 16:16:36 +02:00
meili-bors[bot]	9066a446a3	Merge #4642 4642: Index the _geo fields when changing the setting while there is already documents in the DB r=ManyTheFish a=irevoire # Pull Request ## Related issue Fixes https://github.com/meilisearch/meilisearch/issues/4640 Fixes https://github.com/meilisearch/meilisearch/issues/4628 ## What does this PR do? - Add an integration test that first indexes the document and then changes the settings - Fix `extract_geo_point` by detecting if the `_geo` field has been faceted in this setting change and index all documents Co-authored-by: Tamo <tamo@meilisearch.com> Co-authored-by: ManyTheFish <many@meilisearch.com>	2024-05-21 13:16:11 +00:00
Louis Dureuil	eccbcf5130	Increase index-scheduler test timeouts	2024-05-21 14:59:08 +02:00
Clément Renault	943f8dba0c	Make clippy happy	2024-05-21 14:58:41 +02:00
Clément Renault	1aa8ed9ef7	Make the original sorter optional	2024-05-21 14:53:26 +02:00
ManyTheFish	f762307838	Fix clippy	2024-05-21 13:44:20 +02:00
ManyTheFish	3e94a90722	Fixes	2024-05-21 13:39:46 +02:00
meili-bors[bot]	abe29772db	Merge #4644 4644: Revert "Stream documents" and keep heed+arroy to the latest verion r=Kerollmops a=irevoire Reverts meilisearch/meilisearch#4544 Fixes https://github.com/meilisearch/meilisearch/issues/4641 I didn’t realize that some http clients were not handling chunked http requests like you would expect (if you ask the body, it gives you the body), which made the previous PR breaking. There is no way to provide a good fix to the issue we initially wanted to fix without breaking meilisearch and that’s not planned for now. Co-authored-by: Tamo <irevoire@protonmail.ch> Co-authored-by: Tamo <tamo@meilisearch.com>	2024-05-21 10:21:47 +00:00
Tamo	c9ac7f2e7e	update heed to latest version	2024-05-20 15:19:00 +02:00
Tamo	7e251b43d4	Revert "Stream documents"	2024-05-20 15:09:45 +02:00
Louis Dureuil	9969f7a638	Add test on index-scheduler	2024-05-20 14:44:10 +02:00
Louis Dureuil	b17cb56dee	Test array of vectors	2024-05-20 14:44:10 +02:00
Louis Dureuil	afcd7b9f0c	Test hybrid search with hf embedder	2024-05-20 14:44:10 +02:00
ManyTheFish	fc7e817221	Index geo points based on the settings differences	2024-05-20 12:27:26 +02:00
Tamo	0f78703b85	add a test reproducing the bug	2024-05-20 10:58:08 +02:00
Louis Dureuil	30cf972987	Add test with a dump	2024-05-20 10:36:18 +02:00
Louis Dureuil	d05d49ffd8	Fix tests	2024-05-20 10:36:18 +02:00
Louis Dureuil	0462ebbe58	Don't write an empty _vectors field	2024-05-20 10:36:18 +02:00
Louis Dureuil	2f7a8a4efb	Don't write vectors that weren't autogenerated in document DB	2024-05-20 10:36:18 +02:00
Louis Dureuil	02714ef5ed	Add vectors from vector DB in dump	2024-05-20 10:36:18 +02:00
Louis Dureuil	52d9cb6e5a	Refactor vector indexing - use the parsed_vectors module - only parse `_vectors` once per document, instead of once per embedder per document	2024-05-20 10:36:17 +02:00
Louis Dureuil	261de888b7	Add function to get the embeddings of a document in an index	2024-05-20 10:36:17 +02:00
Louis Dureuil	98c811247e	Add parsed vectors module	2024-05-20 10:25:59 +02:00
meili-bors[bot]	59ecf1cea7	Merge #4544 4544: Stream documents r=curquiza a=irevoire # Pull Request ## Related issue Fixes https://github.com/meilisearch/meilisearch/issues/4383 ### Perf 2M hackernews: main: Time to retrieve: 7s RAM consumption: 2+GiB stream: Time to retrieve: 4.7s RAM consumption: Too small Co-authored-by: Tamo <tamo@meilisearch.com>	2024-05-17 14:49:08 +00:00
Tamo	273c6e8c5c	uses the latest version of heed to get rid of unsafe code	2024-05-16 18:31:32 +02:00
Tamo	897d25780e	update milli to latest version	2024-05-16 18:31:32 +02:00
Tamo	c85d1752dd	keep the same rtxn to compute the filters on the documents and to stream the documents later on	2024-05-16 18:31:32 +02:00
Tamo	8e6ffbfc6f	stream documents	2024-05-16 18:31:32 +02:00
meili-bors[bot]	7c19c072fa	Merge #4631 4631: Split the field id map from the weight of each fields r=Kerollmops a=irevoire # Pull Request ## Related issue Fixes https://github.com/meilisearch/meilisearch/issues/4484 ## What does this PR do? - Make the (internal) searchable fields database always contain the searchable fields (instead of None when the user-defined searchable fields were not defined) - Introduce a new « fieldids_weights_map » that does the mapping between a fieldId and its Weight - Ensure that when two searchable fields are swapped, the field ID map doesn't change anymore (and thus, doesn't re-index) - Uses the weight instead of the order of the searchable fields in the attribute ranking rule at search time - When no searchable attributes are defined, make all their weights equal to zero - When a field is declared as searchable and contains nested fields, all its subfields share the same weight ## Impact on relevancy ### When no searchable attributes are declared When no searchable attributes are declared, all the fields have the same importance instead of randomly giving more importance to the field we've encountered « the most early » in the life of the index. This means that before this PR, send the following json: ```json [ { "id": 0, "name": "kefir", "color": "white" }, { "id": 1, "name": "white", "last name": "spirit" } ] ``` Would make the field `name` more important than the field `color` or `last name`. This means that searching for `white` would make the document `1` automatically higher ranked than the document `0`. After this PR, all the fields have the same weight, and none are considered more important than others. ### When a nested field is made searchable The second behavior change that happened with this PR is in the case you're sending this document, for example: ```json { "id": 0, "name": "tamo", "doggo": { "name": "kefir", "surname": "le kef" }, "catto": "gromez" } ``` Previously, defining the searchable attributes as: `["tamo", "doggo", "catto"]` was actually defining the « real » searchable attributes in the engine as: `["tamo", "doggo", "catto", "doggo.name", "doggo.surname"]`, which means that `doggo.name` and `doggo.surname` were _NOT_ where the user expected them and had completely different weights than `doggo`. In this PR all the weights have been unified, and the « real » searchable fields look like this: ```json [ "tamo", "doggo", "doggo.name", "doggo.surname", "catto"] ^^^^ ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^ Weight 0 Weight 1 Weight 2 Co-authored-by: Tamo <tamo@meilisearch.com>	2024-05-16 09:59:24 +00:00
Tamo	673b6e1dc0	fix a flaky test	2024-05-16 11:28:14 +02:00
Tamo	f2d0a59f1d	when no searchable attributes are defined, makes all the weight equals to zero	2024-05-16 01:06:33 +02:00
Tamo	c78a2fa4f5	rename method and variable around the attributes to search on feature	2024-05-15 18:04:42 +02:00
Tamo	5542f1d9f1	get back to what we were doingb efore in the DB cache and with the restricted field id	2024-05-15 18:00:39 +02:00
Tamo	ad4d8502b3	stops storing the whole fieldids weights map when no searchable are defined	2024-05-15 17:16:10 +02:00
Tamo	7ec4e2a3fb	apply all style review comments	2024-05-15 15:02:26 +02:00
Tamo	9fffb8e83d	make clippy happy	2024-05-14 17:36:32 +02:00
Tamo	caa6a7149a	make the attribute ranking rule use the weights and fix the tests	2024-05-14 17:36:32 +02:00
Tamo	a0082c4df9	add a failing test on the attribute ranking rule	2024-05-14 17:00:02 +02:00
Tamo	b0afe0972e	stop updating the fields ids map when fields are only swapped	2024-05-14 17:00:02 +02:00
Tamo	9ecde41853	add a test on the current behaviour	2024-05-14 17:00:02 +02:00
Tamo	685f452fb2	Fix the indexing of the searchable	2024-05-14 17:00:02 +02:00
Tamo	4e4a1ddff7	gate a test behind the required feature	2024-05-14 17:00:02 +02:00
Tamo	c22460045c	Stops returning an option in the internal searchable fields	2024-05-14 17:00:02 +02:00
meili-bors[bot]	76bb6d565c	Merge #4624 4624: Add "precommands" to benchmark r=dureuill a=dureuill # Pull Request ## Related issue Helps for https://github.com/meilisearch/meilisearch/issues/4493 ## What does this PR do? - Add support for precommands for cargo xtask bench - update benchmark docs - update workload files Co-authored-by: Louis Dureuil <louis@meilisearch.com>	2024-05-13 08:27:56 +00:00
Louis Dureuil	9d3ff11b21	Modify existing workload files to use precommands	2024-05-07 14:03:14 +02:00
Louis Dureuil	43763eb98a	Document precommands	2024-05-07 12:26:22 +02:00
Louis Dureuil	2a0ece814c	Add precommands to workloads	2024-05-07 12:23:36 +02:00