Open source, self-hosted - verified
Chroma (open source)
Chroma is listed under open source, self-hosted (deployment type: open source self-host). Open source; no price. Stated: hybrid and sparse search and built-in embedding. Read from the vendor's own pages on October 18, 2026.
Read from docs.trychroma.com, raw.githubusercontent.com. Docs - License - Compare every vector database
Status
- Availability label
- Not stated
No release-status label is stated on the pages checked.
How Chroma charges
Every price Chroma states, in its own unit, with the quote and where it comes from. Nothing is converted.
Open source; no price.
Free tier and limits
- Free tier
- Not stated
Open-source software; no hosted free tier applies.
- Limits
- Not stated
Not stated for self-hosted Chroma on the pages checked.
Search and indexing
- Max dimensions
- Not stated
A dimension cap for self-hosted Chroma is not stated on the pages checked (the 4,096 cap is a Chroma Cloud quota).
- Distance metrics
l2, cosine, ip
“The default is l2 (squared L2 norm), and other possible values are cosine (cosine similarity), and ip (inner product).”
docs.trychroma.com - Index types
HNSW (single node), SPANN (distributed)
“In Single Node Chroma collections, we use an HNSW (Hierarchical Navigable Small World) index to perform approximate nearest neighbor (ANN) search.”
docs.trychroma.com - Hybrid and sparse search
Yes: dense, sparse and hybrid search
“Vector Search Dense, sparse, and hybrid search.”
docs.trychroma.com - Metadata filtering
Yes
“Metadata Filtering Filter results at query time by metadata conditions.”
docs.trychroma.com - Multi-tenancy
- Not stated
Not described on the pages checked for self-hosted Chroma.
Built-in embedding and reranking
- Built-in embedding
Client-side embedding functions (including a DefaultEmbeddingFunction); not a hosted service
“Embedding functions can be linked to a collection and used whenever you call add, update, upsert or query.”
docs.trychroma.com - Built-in reranking
- Not stated
Not stated on the pages checked.
Deployment, license and SDKs
- Regions and clouds
- Not stated
Self-hosted software; regions do not apply.
- BYOC and self-host
Self-host
“Run it locally, self-host, or use Chroma Cloud for a managed, serverless experience.”
docs.trychroma.com - License
Apache-2.0
“Apache License Version 2.0, January 2004”
raw.githubusercontent.com - SDKs
Python, JavaScript
“pip install chromadb # python client”
raw.githubusercontent.com “for javascript, npm install chromadb!”
raw.githubusercontent.com
Vendor-stated claims
Performance, scale and SLA figures as Chroma states them. Not measured.
No performance, scale or SLA figure is recorded from the pages checked.
Research notes
Single-node Chroma uses HNSW; distributed Chroma and Chroma Cloud use SPANN, and the docs say "We currently don't allow customization or modification of SPANN configuration."
FAQ
How much does Chroma cost?
Open source; no price.
What is the maximum vector dimension in Chroma?
Not stated. A dimension cap for self-hosted Chroma is not stated on the pages checked (the 4,096 cap is a Chroma Cloud quota).
Does Chroma support hybrid search?
Yes: dense, sparse and hybrid search, per the vendor: "Vector Search Dense, sparse, and hybrid search."
Does Chroma have built-in embedding or reranking?
Embedding: Client-side embedding functions (including a DefaultEmbeddingFunction); not a hosted service. Reranking: not stated. Not stated on the pages checked.
Sources
- Source: docs.trychroma.com
- Source: docs.trychroma.com/docs/collections/configure
- Source: docs.trychroma.com/docs/overview/introduction
- Source: docs.trychroma.com/docs/embeddings/embedding-functions
- Source: raw.githubusercontent.com/chroma-core/chroma/main/LICENSE
- Source: raw.githubusercontent.com/chroma-core/chroma/main/README.md
Last verified from each vendor's own docs, pricing pages, READMEs and license files; every quote was string-matched against the page that day. This is a dated snapshot, not a benchmark: prices are shown in each product's own unit and change without notice, so check the linked source before you commit.
Newsletter
Know when a vector database changes its pricing
One email per week. New model releases, retrieval tooling and pricing changes and what they mean for what you build.