Chroma DB
An open-source vector database for AI apps with local deployment, full-text and metadata search, plus a separate managed cloud service.
Chroma DB is a search database for developers building AI apps and agents that need to retrieve relevant information from their data. You can run it locally or on your own infrastructure.
- Semantic and lexical search: Find records by meaning with vector search, or match terms with sparse vectors using BM25 and SPLADE.
- Text matching: Search document text with trigram matching and regular expressions.
- Metadata queries: Filter and facet results by stored attributes, then group and aggregate results by metadata keys; metadata can include arrays of strings, numbers and booleans.
- Dataset forks: Duplicate collections for dataset versioning, A/B tests and rollouts.
The open-source project uses the Apache 2.0 license and has Python and JavaScript/TypeScript clients. Chroma Cloud is a separate hosted service for managed vector, hybrid and full-text search. Its storage uses automatic tiering and caching, and the service scales with data and traffic. Cloud Sync provides serverless ingestion into Chroma Cloud, including web-page indexing and GitHub repository chunking and embedding.
Logotyper
