ChromaDB

Open-source vector database providing semantic, full-text, regex, and metadata search for embedding workloads. Stores and queries large-scale embeddings with support for automatic indexing, metadata filtering, and hybrid search across dense and sparse vectors.

Chroma is Apache 2.0 licensed and widely adopted for AI applications, with millions of monthly downloads across Python, TypeScript, and Rust SDKs. It differentiates from alternatives by unifying vector search, lexical search (BM25, SPLADE), and regex in one system, with storage built on object storage for lower operational cost and automatic data tiering.

Key features:

• Vector similarity search and sparse vector search with BM25 and SPLADE
• Full-text and regex search, metadata filtering and faceted search
• Collection forking for dataset versioning and A/B testing
• Python, TypeScript, and Rust clients, plus CLI for development
• Local, self-hosted, or cloud deployment

Common applications include retrieval-augmented generation, semantic search over documents, knowledge bases for chatbots, and agent memory. Integrates with LangChain and other AI frameworks, and supports embedding models from multiple providers.