sqlite-vec/README.md

161 lines
9.2 KiB
Markdown
Raw Normal View History

2024-04-20 13:38:58 -07:00
# `sqlite-vec`
2024-08-01 02:45:36 -07:00
[![](https://dcbadge.vercel.app/api/server/VCtQ8cGhUs)](https://discord.gg/Ve7WeCJFXk)
2024-06-12 00:08:53 -07:00
2024-04-20 13:38:58 -07:00
An extremely small, "fast enough" vector search SQLite extension that runs
2024-08-01 02:45:36 -07:00
anywhere! A successor to [`sqlite-vss`](https://github.com/asg017/sqlite-vss)
2024-04-20 13:38:58 -07:00
2024-06-12 00:13:56 -07:00
<!-- deno-fmt-ignore-start -->
> [!IMPORTANT]
2024-08-01 02:45:36 -07:00
> _`sqlite-vec` is a pre-v1, so expect breaking changes!_
2024-06-12 00:13:56 -07:00
<!-- deno-fmt-ignore-end -->
2024-04-20 13:38:58 -07:00
- Store and query float, int8, and binary vectors in `vec0` virtual tables
2024-06-12 00:13:56 -07:00
- Written in pure C, no dependencies, runs anywhere SQLite runs
(Linux/MacOS/Windows, in the browser with WASM, Raspberry Pis, etc.)
2025-01-18 10:15:09 -08:00
- Store non-vector data in metadata, auxiliary, or partition key columns
2024-04-20 13:38:58 -07:00
2024-06-12 00:08:53 -07:00
<p align="center">
2024-06-25 08:41:42 -07:00
<a href="https://hacks.mozilla.org/2024/06/sponsoring-sqlite-vec-to-enable-more-powerful-local-ai-applications/">
2024-06-12 13:48:34 -07:00
<picture>
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/mozilla.dark.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/mozilla.svg">
<img alt="Mozilla Builders logo" width=400>
</picture>
2024-06-25 08:41:42 -07:00
</a>
2024-06-12 00:08:53 -07:00
</p>
<p align="center">
<i>
<code>sqlite-vec</code> is a
2024-06-25 08:41:42 -07:00
<a href="https://hacks.mozilla.org/2024/06/sponsoring-sqlite-vec-to-enable-more-powerful-local-ai-applications/">Mozilla Builders project</a>,
2024-06-12 00:08:53 -07:00
with additional sponsorship from
<a href="https://fly.io/"><img width=14px src="./.github/logos/flyio.small.ico"/> Fly.io </a>,
2025-01-18 10:15:09 -08:00
<a href="https://tur.so/sqlite-vec"><img width=14px src="./.github/logos/turso.small.ico"/> Turso</a>,
<a href="https://sqlitecloud.io/"><img width=14px src="./.github/logos/sqlitecloud.small.svg"/> SQLite Cloud</a>, and
<a href="https://dcspark.io/"><img width=14px src="./.github/logos/dcspark.small.svg"/> DcSpark</a>.
2024-06-12 00:08:53 -07:00
See <a href="#sponsors">the Sponsors section</a> for more details.
</i>
</p>
2024-06-22 16:46:33 -07:00
## Installing
2024-08-08 02:10:56 +02:00
See [Installing `sqlite-vec`](https://alexgarcia.xyz/sqlite-vec/installation.html)
2024-06-22 16:46:33 -07:00
for more details.
| Language | Install | More Info | |
| -------------- | ---------------------------------------------------- | ------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Python | `pip install sqlite-vec` | [`sqlite-vec` with Python](https://alexgarcia.xyz/sqlite-vec/python.html) | [![PyPI](https://img.shields.io/pypi/v/sqlite-vec.svg?color=blue&logo=python&logoColor=white)](https://pypi.org/project/sqlite-vec/) |
| Node.js | `npm install sqlite-vec` | [`sqlite-vec` with Node.js](https://alexgarcia.xyz/sqlite-vec/js.html) | [![npm](https://img.shields.io/npm/v/sqlite-vec.svg?color=green&logo=nodedotjs&logoColor=white)](https://www.npmjs.com/package/sqlite-vec) |
2024-08-01 02:45:36 -07:00
| Ruby | `gem install sqlite-vec` | [`sqlite-vec` with Ruby](https://alexgarcia.xyz/sqlite-vec/ruby.html) | ![Gem](https://img.shields.io/gem/v/sqlite-vec?color=red&logo=rubygems&logoColor=white) |
| Go | `go get -u github.com/asg017/sqlite-vec/bindings/go` | [`sqlite-vec` with Go](https://alexgarcia.xyz/sqlite-vec/go.html) | [![Go Reference](https://pkg.go.dev/badge/github.com/asg017/sqlite-vec-go-bindings/cgo.svg)](https://pkg.go.dev/github.com/asg017/asg017/sqlite-vec-go-bindings/cgo) |
2024-06-22 16:46:33 -07:00
| Rust | `cargo add sqlite-vec` | [`sqlite-vec` with Rust](https://alexgarcia.xyz/sqlite-vec/rust.html) | [![Crates.io](https://img.shields.io/crates/v/sqlite-vec?logo=rust)](https://crates.io/crates/sqlite-vec) |
| Datasette | `datasette install datasette-sqlite-vec` | [`sqlite-vec` with Datasette](https://alexgarcia.xyz/sqlite-vec/datasette.html) | [![Datasette](https://img.shields.io/pypi/v/datasette-sqlite-vec.svg?color=B6B6D9&label=Datasette+plugin&logoColor=white&logo=python)](https://datasette.io/plugins/datasette-sqlite-vec) |
| rqlite | `rqlited -extensions-path=sqlite-vec.tar.gz` | [`sqlite-vec` with rqlite](https://alexgarcia.xyz/sqlite-vec/rqlite.html) | [![rqlite](https://img.shields.io/badge/rqlite-sqlite_extensions-blue)](https://rqlite.io/docs/guides/extensions/) |
2024-06-22 16:46:33 -07:00
| `sqlite-utils` | `sqlite-utils install sqlite-utils-sqlite-vec` | [`sqlite-vec` with sqlite-utils](https://alexgarcia.xyz/sqlite-vec/sqlite-utils.html) | [![sqlite-utils](https://img.shields.io/pypi/v/sqlite-utils-sqlite-vec.svg?color=B6B6D9&label=sqlite-utils+plugin&logoColor=white&logo=python)](https://datasette.io/plugins/datasette-sqlite-vec) |
| Github Release | | | ![GitHub tag (latest SemVer pre-release)](https://img.shields.io/github/v/tag/asg017/sqlite-vec?color=lightgrey&include_prereleases&label=Github+release&logo=github) |
2024-04-20 13:38:58 -07:00
## Sample usage
```sql
.load ./vec0
create virtual table vec_examples using vec0(
sample_embedding float[8]
);
-- vectors can be provided as JSON or in a compact binary format
insert into vec_examples(rowid, sample_embedding)
values
(1, '[-0.200, 0.250, 0.341, -0.211, 0.645, 0.935, -0.316, -0.924]'),
(2, '[0.443, -0.501, 0.355, -0.771, 0.707, -0.708, -0.185, 0.362]'),
(3, '[0.716, -0.927, 0.134, 0.052, -0.669, 0.793, -0.634, -0.162]'),
(4, '[-0.710, 0.330, 0.656, 0.041, -0.990, 0.726, 0.385, -0.958]');
2024-08-01 02:45:36 -07:00
-- KNN style query
2024-04-20 13:38:58 -07:00
select
rowid,
distance
from vec_examples
where sample_embedding match '[0.890, 0.544, 0.825, 0.961, 0.358, 0.0196, 0.521, 0.175]'
order by distance
limit 2;
/*
┌───────┬──────────────────┐
│ rowid │ distance │
├───────┼──────────────────┤
│ 2 │ 2.38687372207642 │
│ 1 │ 2.38978505134583 │
└───────┴──────────────────┘
*/
```
2024-06-12 10:45:59 -07:00
## Sponsors
2024-06-12 00:08:53 -07:00
2024-06-12 00:13:56 -07:00
Development of `sqlite-vec` is supported by multiple generous sponsors! Mozilla
is the main sponsor through the new Builders project.
2024-06-12 00:08:53 -07:00
<p align="center">
2024-06-25 08:41:42 -07:00
<a href="https://hacks.mozilla.org/2024/06/sponsoring-sqlite-vec-to-enable-more-powerful-local-ai-applications/">
2024-06-12 13:48:02 -07:00
<picture>
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/mozilla.dark.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/mozilla.svg">
<img alt="Mozilla Builders logo" width=400>
</picture>
2024-06-25 08:41:42 -07:00
</a>
2024-06-12 00:08:53 -07:00
</p>
`sqlite-vec` is also sponsored by the following companies:
2024-06-12 00:23:45 -07:00
<a href="https://fly.io/">
2024-06-12 00:27:15 -07:00
<picture>
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/flyio.dark.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/flyio.svg">
2025-01-18 10:15:09 -08:00
<img alt="Fly.io logo" src="./.github/logos/flyio.svg" width="48%">
2024-06-12 00:27:15 -07:00
</picture>
2024-06-12 00:23:45 -07:00
</a>
2024-06-12 00:27:15 -07:00
2024-08-05 14:14:40 -04:00
<a href="https://tur.so/sqlite-vec">
2024-06-12 00:27:15 -07:00
<picture>
2024-06-12 00:30:10 -07:00
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/turso.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/turso.svg">
2025-01-18 10:15:09 -08:00
<img alt="Turso logo" src="./.github/logos/turso.svg" width="48%">
2024-06-12 00:27:15 -07:00
</picture>
2024-06-12 00:23:45 -07:00
</a>
2024-06-12 00:27:15 -07:00
2024-06-12 00:23:45 -07:00
<a href="https://sqlitecloud.io/">
2024-06-12 00:27:15 -07:00
<picture>
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/sqlitecloud.dark.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/sqlitecloud.svg">
2025-01-18 10:15:09 -08:00
<img alt="SQLite Cloud logo" src="./.github/logos/flyio.svg" width="48%">
</picture>
</a>
<a href="https://dcspark.io/">
<picture>
2025-01-21 13:16:29 -08:00
<source media="(prefers-color-scheme: dark)" srcset="./.github/logos/dcspark.dark.svg">
<source media="(prefers-color-scheme: light)" srcset="./.github/logos/dcspark.svg">
2025-01-18 10:15:09 -08:00
<img alt="DcSpark logo" src="./.github/logos/dcspark.svg" width="48%">
2024-06-12 00:27:15 -07:00
</picture>
2024-06-12 00:13:56 -07:00
</a>
2024-06-12 00:08:53 -07:00
2024-06-12 00:13:56 -07:00
As well as multiple individual supporters on
[Github sponsors](https://github.com/sponsors/asg017/)!
2024-06-12 00:08:53 -07:00
2024-06-12 00:13:56 -07:00
If your company interested in sponsoring `sqlite-vec` development, send me an
email to get more info: https://alexgarcia.xyz
2024-04-20 13:38:58 -07:00
2024-06-12 00:08:53 -07:00
## See Also
2024-04-20 13:38:58 -07:00
2024-06-12 00:13:56 -07:00
- [**`sqlite-ecosystem`**](https://github.com/asg017/sqlite-ecosystem), Maybe
more 3rd party SQLite extensions I've developed
- [**`sqlite-rembed`**](https://github.com/asg017/sqlite-rembed), Generate text
embeddings from remote APIs like OpenAI/Nomic/Ollama, meant for testing and
SQL scripts
- [**`sqlite-lembed`**](https://github.com/asg017/sqlite-lembed), Generate text
embeddings locally from embedding models in the `.gguf` format