Chroma review

Q: How do I install Chroma?

Run pip install chromadb. It needs Node.js >=20. The install section above has the other options from the project's README.

Chroma is open-source search infrastructure for AI, offering vector, full-text, regex, and metadata search through Python and JavaScript clients.

Updated Nov 1, 2026. Review written Oct 22, 2026 from the project's own sources; usage data refreshed every Monday.

TL;DR

Chroma is an open-source, Apache 2.0 search database for AI applications, with Python and JavaScript clients. It suits developers building retrieval and agent features who want a small API that handles embedding and indexing. It runs in-memory, in client-server mode, or as the hosted Chroma Cloud.

29.6kGitHub stars
+51new stars, week of Oct 25
~2.3Mdownloads a week, +10% vs 4-week avg
#28in RAG, data and memory this week

Chroma at a glance

Category: RAG, data and memory
Maintained by: Chroma (organization, @chroma-core on GitHub)
License: Apache 2.0
Language: Rust, Python, TypeScript
Repository: github.com/chroma-core/chroma
Website: www.trychroma.com
npm package: chromadb 3.5.0
PyPI package: chromadb 1.5.9
Latest release: chromadb 3.5.0 on npm, Jun 30, 2026
First published: Oct 5, 2022

What Chroma does

Chroma stores documents in collections and handles tokenization, embedding, and indexing automatically, or accepts your own embeddings.
Chroma queries collections for the most similar results to a query text and can also get documents by id.
Chroma filters results on metadata fields and on document contents, such as a $contains string match.
Chroma supports sparse vector search for lexical ranking with BM25 and SPLADE.
Chroma offers full-text search with trigram and regex matching, plus faceted metadata search.
Chroma supports forking for dataset versioning, A/B testing, and roll-outs.
Chroma includes command-line tools for development.

How it works

Chroma installs as the chromadb package for Python or npm. You can run it in-memory for prototyping, add persistence, or start a server with chroma run for client-server mode. The hosted Chroma Cloud provides serverless search built on object storage such as S3 or GCS, with memory and SSD caching for hot and warm data.

How to install and start using Chroma

With pip

pip install chromadb

With npm

npm install chromadb

Run in client-server mode

chroma run --path /chroma_db_path

Requirements

Python 3.9 or later for the Python package.
Node.js 20 or later for the JavaScript package.
A Chroma Cloud account for the hosted service.

First steps

Create a client with chromadb.Client() to run Chroma in-memory for prototyping.
Create a collection with client.create_collection and a name such as "all-my-documents".
Add documents with collection.add, passing documents, optional metadatas, and a unique id for each document.
Search with collection.query using query_texts and n_results, optionally adding where or where_document filters.

Who Chroma is for

Developers prototyping retrieval for AI apps who want a small API with automatic embedding.
Teams that need vector, full-text, regex, and metadata search in one system.
Python and TypeScript developers building search features for AI agents.
Teams that want a hosted, serverless option instead of running search infrastructure themselves.

Limitations to know

Chroma describes itself as a rapidly developing project, so APIs may change; a row-based API is listed as coming soon.
Chroma Cloud lists per-collection limits, including 5M records per collection and 1M collections per database.
Chroma Cloud cold queries are slower than warm ones in the homepage's published latency figures.

Is Chroma growing? Usage data

In the week of Oct 25–31, 2026, Chroma ranked #248 on the Devtoolwatch Board and #28 in RAG, data and memory, with a momentum score of 38 out of 100. Its main package had about 2.3M downloads that week, +10% against its own 4-week average.

New GitHub stars per week for the last 12 weeks, from GitHub's stargazer history (today's stargazers by the week they starred).

Show the weekly numbers

New GitHub stars per week for Chroma
Week of	New stars
Aug 9, 2026	84
Aug 16, 2026	80
Aug 23, 2026	62
Aug 30, 2026	64
Sep 6, 2026	58
Sep 13, 2026	57
Sep 20, 2026	66
Sep 27, 2026	52
Oct 4, 2026	58
Oct 11, 2026	60
Oct 18, 2026	62
Oct 25, 2026	51

On Hacker News

Context Rot: How increasing input tokens impacts LLM performance, 260 points and 59 comments (Jul 14, 2025)
Show HN: Chroma Cloud – serverless search database for AI, 93 points and 40 comments (Aug 18, 2025)
Chroma Context-1: Training a Self-Editing Search Agent, 71 points and 8 comments (Mar 26, 2026)
Wal3: A Write-Ahead Log for Chroma, Built on Object Storage, 66 points and 8 comments (Sep 4, 2025)
Designing a Query Execution Engine, 41 points and 0 comments (Oct 30, 2024)

The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.

Chroma alternatives

Other RAG, data and memory tools on the Board in the week of Oct 25–31, 2026, by momentum:

RAG, data and memory ranked by momentum, the week of Oct 25–31, 2026
Rank	Tool	Stars	Downloads/wk
#1	XERJXERJ is a local-first search engine and MCP server for AI agents, packaged as one Rust binary with an Elasticsearch-compatible API.	17.9k	–
#2	MemvidMemvid is a single-file memory layer for AI agents that packages data, embeddings, search indexes, and metadata into one portable .mv2 file.	17.3k	–
#3	LEANNLEANN is a Python vector database and RAG library that recomputes embeddings on demand to keep indexes small on personal devices.	13.7k	–
#4	GraphifyGraphify is an AI coding assistant skill that turns a codebase, with its docs, PDFs, images and videos, into a queryable knowledge graph.	140k	480.7k
#5	MarkItDownMarkItDown is a lightweight Python utility from Microsoft that converts PDFs, Office documents, images, audio and other files to…	193.2k	4.7M
#6	FirecrawlFirecrawl is an API and open source project to search, scrape, crawl, and interact with web pages, returning LLM-ready markdown or JSON…	197.4k	2.9M

All RAG, data and memory tools, ranked

Recent Chroma releases

latest (pre-release), Apr 1, 2025
1.5.9, May 5, 2026
cli-1.4.4, May 5, 2026
foundation-cli-v0.1.0-alpha.3 (pre-release), May 1, 2026
1.5.8, Apr 16, 2026

Chroma FAQ

Is Chroma free and open source?

Yes. Chroma's source code is on GitHub (chroma-core/chroma) under the Apache 2.0 license, so you can use it for free. Its website also links to a pricing page (https://www.trychroma.com/pricing) for any paid plans.

How do I install Chroma?

Run pip install chromadb. It needs Node.js >=20. The install section above has the other options from the project's README.

Is Chroma still maintained?

Yes. The chroma-core/chroma repository was last updated on Oct 22, 2026, and the latest release, chromadb 3.5.0 on npm, came out on Jun 30, 2026.

How popular is Chroma?

Chroma has 29.6k GitHub stars as of Nov 1, 2026. It gained 51 new stars in the week of Oct 25–31, 2026, against a 12-week average of 63 a week. Its main package had about 2.3M downloads that week, +10% against its own 4-week average. That made it #28 among RAG, data and memory tools on the Devtoolwatch Board.

Chroma vs Qdrant: how do they compare?

Chroma and Qdrant are both tracked by Devtoolwatch as RAG, data and memory tools. Chroma has 29.6k GitHub stars and Qdrant has 35.4k. In the week of Oct 25–31, 2026, Chroma's main package had 2.3M downloads and Qdrant's had 4.5M. Qdrant is an open-source vector similarity search engine and vector database written in Rust, with filtering on JSON payloads. Qdrant review.

Chroma vs pgvector: how do they compare?

Chroma and pgvector are both tracked by Devtoolwatch as RAG, data and memory tools. Chroma has 29.6k GitHub stars and pgvector has 23.5k. In the week of Oct 25–31, 2026, Chroma's main package had 2.3M downloads and pgvector's had 9.4M. pgvector is an open-source Postgres extension for exact and approximate vector similarity search, storing vectors alongside the rest of your data. pgvector review.

What are the best alternatives to Chroma?

Other RAG, data and memory tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: XERJ (#1), Memvid (#2), LEANN (#3), Graphify (#4), MarkItDown (#5). The alternatives table above has their numbers.

Is ChromaDB free?

Chroma is open source under the Apache 2.0 license. The hosted Chroma Cloud is a separate service, and the README says new users get $5 of free credits.

Can Chroma create embeddings for me?

Chroma handles tokenization, embedding, and indexing automatically when you add documents. You can also skip that step and supply your own embeddings.

Does Chroma support keyword or full-text search, not just vectors?

Chroma supports sparse vector search with BM25 and SPLADE, full-text search with trigram and regex matching, and metadata filtering alongside vector similarity search.

How often does Chroma release new versions?

Chroma releases new tagged versions of its pypi and npm packages on Mondays. Hotfixes ship at any time during the week.

Can I run Chroma as a server?

Chroma runs in client-server mode with the chroma run command and a data path. It can also run in-memory inside your application for prototyping.

Sources

Chroma README badge

Maintainers can show the tool's weekly Board rank in their README:

[![Devtoolwatch](https://index.devtoolwatch.com/badge/chroma.svg)](https://index.devtoolwatch.com/tools/chroma/)

How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].