Introducing Chroma Cloud: an open-source serverless search database that is fast, cost-effective, scalable, and reliable.

Chroma Cloud abstracts away all the complexity of the database so you can focus on your application and your users. No VMs to manage or configurations to tune. Watch the demo.

Chroma Cloud is backed by Distributed Chroma, an Apache 2.0 project. Learn about the lifecycle of read and write requests, and how Chroma delivers low latency search on top of object storage.

AI applications need many small indexes. Chroma Cloud calls these “Collections”, and can horizontally scale to billions. Leverage full-text, metadata, vector, and regex search. Use copy-on-write forking to cheaply version your data.

Chroma Cloud is live in production with thousands of users, from leading AI agents to fast growing startups.

Only pay for what you use: $0 / month + $5 in free credits $2.50 / GiB written $0.33 / GiB / mo stored $0.0075 / TiB queried + $0.09 GiB / returned

We are committed to building Chroma as a ubiquitous open source standard for search and retrieval. Chroma is downloaded over 5 million times a month. Join our community.

Try it for yourself - create a database in 30 seconds.

Congrats team! Excited for more folks to experience Chroma cloud -- it's been a great experience for us

I love chroma cloud

love the distored analog synths 🎶

The vibes are too good.

Congrats on the launch, excited to see this live! 🎉

Local chroma has been awesome. This will be great for scaling!

C o z y vibes

Woohoo! Awesome!

have loved using it so far 🫶🫶

congrats on the launch Jeff and team!

Nostalgic, iconic and immedietly applicable The ad is a vibe

This sounds promising.

love it.

🔥

it's good but a lot of annoying limitations that pinecone doesn't have, particularly around metadata storing and filtering. I'd prefer developing for pinecone, but since their pricing change, chroma letting you pay what you use and not a flat $50/m my clients are happier with it.

Awesome ad

Congrats, Jeff and team Chroma!

cool concept. serverless is a game changer for scaling without the usual headaches. curious about the performance in high-load scenarios.

congrats team 🎉

Let's goo congrats on the launch

Serverless vector search solves the infrastructure problem for RAG. The agent behavior problem remains: does the agent USE the retrieved context correctly? We're seeing: - Agents retrieve relevant docs but extract wrong information - Agents ignore critical context in favor of cached knowledge - Agents hallucinate even with perfect retrieval results Testing how agents process and act on Chroma's results could prevent these production failures.

Skimr is using Chroma Cloud. We can’t recommend it more 💪💪💪

9

This is the kind of thinking I vibe with—chat soon?☄️💡