Chroma

Chroma

Rust-based vector database for AI — serverless cloud with hybrid and full-text search for embeddings, metadata, and RAG

Open source alternative to:Pinecone

Chroma is an open-source vector database with 28k+ GitHub stars — a Pinecone alternative rebuilt in Rust for high-performance AI search and retrieval. Serverless cloud with hybrid and full-text search capabilities.

Key features

Vector search

  • High-performance similarity search with HNSW indexing
  • Hybrid search combining vectors, keywords, and metadata
  • Full-text search with BM25 ranking
  • Metadata filtering with rich query language
  • Multi-tenancy support out of the box

Database capabilities

  • Rust-based engine for memory efficiency and speed
  • Serverless cloud with automatic scaling
  • ACID transactions and data durability
  • Distributed architecture for high availability
  • Multi-vector support per document

Developer experience

  • Simple Python, JavaScript/TypeScript, and Rust SDKs
  • Seamless integration with LangChain, LlamaIndex, and OpenAI
  • Automatic embedding generation with built-in models
  • REST API for language-agnostic access

At a glance

LicenseApache-2.0
StackRust, Python, JavaScript
Self-hostedYes — Chroma OSS
CloudChroma Cloud (serverless)
APIPython, JS/TS, Rust, REST

Self-hosting

pip install chromadb

Chroma can be embedded directly in your application or run as a standalone server. The cloud version provides fully managed infrastructure with automatic scaling.

Screenshots

Chroma screenshot 1

Category

Developer Tools

Tags

aivector-databaseembeddingsrag