asg017/sqlite-vec

Health v2 · maintenance only

Adoption not part of the score

8168 stars · 352 forks observed · 2026-10-08

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

sqlite-vec is a vector search extension for SQLite written in pure C with no dependencies, supporting float, int8, and binary vectors in vec0 virtual tables. It runs anywhere SQLite runs, including browsers via WASM and embedded devices like Raspberry Pis.

Use cases

• add vector search to a local SQLite database
• store and query embeddings for a RAG application
• run semantic search entirely on-device without a server
• build a local AI app with vector similarity search
• search binary or int8 embeddings in SQLite
• prototype vector search without deploying a vector database

When to choose

• you want lightweight, dependency-free vector search embedded in SQLite
• you need local-first or on-device semantic search
• your dataset is small-to-medium and 'fast enough' beats heavyweight infrastructure
• you need portability across platforms including WASM and embedded devices

When to avoid

• you need billion-scale, GPU-accelerated vector search at high throughput
• you require a stable API - the project is pre-v1 with breaking changes
• you need advanced ANN indexes like HNSW with heavy tuning options

Facets

library · maturity active

vector-database search-engine database databases machine-learning developer-tools windows wasm python cli cross-platform sqlite-extension embeddings vector-search local-ai rag search linux macos nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="asg017/sqlite-vec")