Chroma
Limited native hybrid (users build RRF custom). Fast Rust core (v2.5).
At a glance
- Type
- Embedded vector DB (Rust core)
- Tier
- T1
- Section
- Vector-database infrastructure
- Created
- 2022-10
- Latest release
- 1.5.9 2026-05-05
- License
- Apache-2.0
- GitHub
- 27.8k★ +125/mo Rust
- Pricing
- Free + paid
- Funding
- $18M total $75M val Seed · 2023-04
Taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- long-term
- update
- overwrite
- unit
- chunk
- governance
- inspectable
- conflict
- overwrite
When to use
Optimised for: low-latency similarity search + scale
Anti-fit: not for relational / graph-heavy queries (vector-first by design)
Pros & cons
Pros
Lowest-friction local vector store — embedded mode means zero ops for prototyping; popular default in agent tutorials.
Cons
Production scale-out story is less mature than Weaviate / Qdrant; managed cloud (Chroma Cloud) is newer.
Claims & capabilities
26k+ GitHub★ (some sources cite 25.6k); used in 90k+ open-source codebases on GitHub; downloaded 11M+ times monthly; Chroma Cloud GA August 2025; 4x faster writes/queries vs original Python after 2025 Rust rewrite; included in Zilliz VectorDBBench standard cross-engine benchmarks
Technical surface
- API surface
- REST, SDK: Python, JS/TS
- Backend storage
- SQLite (local) / custom cloud backend
- Deployment
- Both
- Embedding model
- multiple supported
- Multi-tenancy
- Multi-tenant indexes with billions of vectors; AWS PrivateLink; BYOC (Bring Your Own Cloud) for hard isolation
- MCP
- via official adapter — chroma-mcp
- A2A
- no Google A2A (Agent2Agent) integration documented as of 2026-05.
- OpenTelemetry
- via OpenTelemetry instrumentation
Compare Chroma with…
Similar systems
Other vector-database infrastructure in the catalog, ranked by inbound references.
- Qdrant T1
Distribution-Based Score Fusion + RRF. Sparse vectors native; filtering via ANN graph modification.
- pgvector T1
Stores embeddings alongside relational + full-text data. HNSW + IVFFlat ANN indexes. Used as agent conversation memory via LangChain + MCP. Foundation of Supabase AI and many self-hosted RAG stacks.
- Pinecone T1
Managed vector DB. Cascading sparse + dense + rerank pipeline; pinecone-rerank-v0 .
- LanceDB T1
Embedded vector DB (Arrow columnar). RRF reranker. Petabyte-scale on disk.
- MongoDB Atlas Vector Search T1
Agent memory store for both short-term (document) and long-term (vector). LangGraph checkpointer for stateful agents. Vector search extended to Community Edition (Sept 2025).
- Milvus T1
Multi-vector columns (10 simultaneous). Native hybrid search (v2.5). CAGRA + Vamana GPU/CPU (v2.6).
Related systems
Referenced by (3)
- CrewAI Memory depends on at runtime — . Short-term memory backed by ChromaDB + RAG; long-term + entity memory layers built in. Mature mu
- MemPalace builds on — Independent analysis suggests the score is driven by verbatim storage + ChromaDB defaults rather than the palace structure.
- Superpowers episodic-memory plugin integrates with — SQLite + FTS5 with optional Chroma vector