ChromaDB
Open-source embedding database for AI applications
View on GitHubWhat it solves
Persisting, querying, and filtering vectors without standing up a full database stack.
Overview
ChromaDB is an open-source vector database designed specifically for AI applications, embedding-native, with first-class Python and JavaScript clients. Runs in-memory, on disk, or as a separate server. Designed for fast prototyping; the same API that works locally runs at scale.
Key facts
- Language
- Python
- License
- Apache 2.0
- Maturity
- Community canon
- Maintainer
- chroma-core
- Reviewed
- 2026-06-11
Where to start
The getting-started covers the collection model (the core abstraction) in five minutes.
docs/getting-started.mdTechnologies
More in this domain
LlamaIndex
Data framework for connecting LLMs to custom data sources
Turning heterogeneous data sources into LLM-queryable indices without writing custom parsers and retrieval pipelines for each format.
pgvector
Vector similarity search extension for PostgreSQL
Infrastructure overhead for teams already on Postgres: semantic search without adding a separate vector database to operate.