Skip to main content
Chroma gives you everything you need for retrieval: store embeddings with metadata, search with dense and sparse vectors, filter by metadata, and retrieve across text, images, and more.

​
What Chroma Offers

Document Storage

Store documents and metadata.

Embeddings

Use any embedding model. OpenAI, Cohere, Hugging Face, sentence-transformers, and more.

Vector Search

Dense, sparse, and hybrid search. Query by similarity and combine multiple search strategies.

Full-Text & Regex Search

Keyword and regex search over your data without embeddings.

Metadata Filtering

Filter results at query time by metadata conditions.

Multi-Modal Retrieval

Index and search images, audio, and other modalities alongside text.

​
Quickstart

Getting Started with the Chroma SDK

Create a self-hosted or cloud database and add data to it using the Chroma SDK.

Create a Chroma Cloud Database

Create a scalable, zero-ops Chroma Cloud database to store your AI data.

​
Example Projects

Agentic Search

Build agents that iteratively search and refine results for complex queries.

Code Search

Index codebases to power coding agents using AST-aware chunking.

​
Open Source

Chroma is licensed under Apache 2.0. Run it locally, self-host, or use Chroma Cloud for a managed, serverless experience.