sqlite-vec

What it is and what it does

sqlite-vec is a SQLite extension that brings vector search and similarity operations to SQLite databases. It allows you to store embeddings and perform vector queries directly in SQL, eliminating the need for a separate vector database for simple use cases. The package is distributed as platform-specific wheels for macOS (both Intel and ARM), Linux (x86_64 and aarch64), and Windows, with no additional runtime dependencies beyond SQLite itself.

The extension is actively maintained and has achieved significant adoption, ranking in the top 5000 PyPI packages by downloads. It's designed for developers who want vector search capabilities without introducing external dependencies or complex infrastructure, keeping everything within the familiar SQLite ecosystem.

Use it for:

• Add semantic search to SQLite-backed applications without deploying a separate vector database
• Store and query embeddings from language models or image encoders within existing SQLite schemas
• Build recommendation systems that use vector similarity on small to medium datasets
• Prototype AI features quickly by embedding vector operations directly in SQL queries

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

sqlite-vec adds vector search capabilities to SQLite, enabling similarity queries and vector operations directly within SQLite databases.

Yes, if you need vector search in SQLite and accept the platform-wheel installation model. The package is actively maintained, permissively licensed, has no runtime dependencies, and is popular enough to indicate real-world usage. The main uncertainty is the unspecified Python version requirement and lack of documentation—verify compatibility with your Python version before committing.

Install

sqlite-vec on PyPI

pip install sqlite-vec

uv add sqlite-vec

poetry add sqlite-vec

Installing sqlite-vec

Before you install

Medium install friction due to platform-specific wheels (macOS x86_64/arm64, Linux aarch64/x86_64, Windows amd64), but no runtime dependencies. Active maintenance with recent release activity.

License in practice

Dual-licensed under MIT and Apache 2.0, both permissive licenses allowing commercial and private use with minimal restrictions.

Quickstart

pip install sqlite-vec
import sqlite_vec
# Vector search operations performed within SQLite queries

Requires a platform-specific wheel match (macOS, Linux, or Windows); Python version requirement unspecified in package metadata.

Verify before relying

• Whether the package requires a specific Python version (requires_python is unspecified in metadata)
• What vector operations and indexing strategies are supported
• Performance characteristics and scalability limits for vector search
• Integration requirements with existing SQLite installations

Package facts

Evidence: sqlite_vec-0.1.9-py3-none-macosx_10_6_x86_64.whl; sqlite_vec-0.1.9-py3-none-macosx_11_0_arm64.whl; sqlite_vec-0.1.9-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl; sqlite_vec-0.1.9-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.manylinux1_x86_64.whl; sqlite_vec-0.1.9-py3-none-win_amd64.whl

Tags

More Database packages