Chroma

Data infrastructure for AI

Apache 2.0

[Chroma @trychroma]

We think memory is the largest unsolved problem in AI today

Introducing Foundation: Chroma's solution to memory

[Jeff Huber @jeffreyhuber]

I’ve been looking forward to today for 3 years

Today we’re announcing Foundation - Chroma’s solution to memory

Our research preview of this technology builds self-improving memory from your agent sessions. Try it out at https://t.co/DpynS0vnSd

[Amjad Masad @amasad]

Thrilled to welcome Jeff and the Chroma team to Replit!

They’re pioneers in AI. They were building RAG before it was even a thing!

At Replit, they’ll help solve one of the hardest problems in AI: giving agents the context and memory they need to do real work.

Massive step toward the self-driving company.

[Chroma @trychroma]

Chroma is joining Replit!

Together, we will build the context layer for the self-driving company.

https://t.co/c4pTGzgcTz

[Chroma @trychroma]

We invented a new concurrency control protocol, named Fission, for agents swarms to update shared data for Foundation.

https://t.co/lXc5aAmY15

and includes a section on "why not Git"

[Chroma @trychroma]

a new emoji - from our slack to yours 🫶

[Jeff Huber @jeffreyhuber]

if you are an engineer and want to build the best memory system on earth - send me a DM

[Jeff Huber @jeffreyhuber]

I’ve been looking forward to today for 3 years

Today we’re announcing Foundation - Chroma’s solution to memory

Our research preview of this technology builds self-improving memory from your agent sessions. Try it out at https://t.co/DpynS0vnSd

[Jeff Huber @jeffreyhuber]

seeing a lot of love for WAL+S3

read how Chroma does it - open-source

https://t.co/TYQnVSpr0d

[Chroma @trychroma]

@DanJCleary Neat! What do you attribute the decrease in performance at 1k tokens to with opus 4.8?

[Chroma @trychroma]

We’re excited to share our latest research led by @patpcj and @HammadTime.

Frontier level agentic search performance at 20B parameters.

[Patrick Jiang @patpcj]

Introducing Harness-1, a 20B search agent trained with a state-externalizing harness.

> frontier-level long-horizon search, rivaling Opus-4.6 and outperforming GPT-5.4

> Context-1-level cost and latency

> externalizes candidates, evidence, verification, and search history

> open-source

[Chroma @trychroma]

Built on Chroma

https://t.co/0iF3CJYUeK

[Chroma @trychroma]

Get your agent working on search and retrieval for you.

https://t.co/AsZrffjVJ2

[Stripe @stripe]

Vibedeploy with Stripe Projects, now available to everyone: https://t.co/fZqyuRojGD.

[Chroma @trychroma]

Chroma Cloud is now available via Stripe Projects!

[Chroma @trychroma]

Reranking is a critical part of a high quality search system.

Build an intuition for how it works and how to think about integrating reranking into your application. 👇

[Chroma @trychroma]

Learn more:

https://t.co/Vnssm5kKAT

[Chroma @trychroma]

New:

EU region support for Chroma Cloud.

Create databases in GCP’s europe-west1. Available now on all plans.

[Chroma @trychroma]

Join us tomorrow night!

[arlan rakh @arlanr]

Introducing No Latency Demo Night

private event — [@nozomioai x linkup x @trychroma]

apr 9, sf

bringing together the most cracked builders for live demos + raw technical creativity.

build anything with chroma, linkup, or nozomio:

agents, retrieval systems, workflows, weird hacks

top 3 demo live
$500 prize + credits + high-signal feedback

apply 24h in advance with a short pitch

https://t.co/n4vnL0GYEf

[Chroma @trychroma]

@techsaleshackz 👋

https://t.co/Nlc2jmYr20

[Chroma @trychroma]

Working with sparse vectors via Schema():

https://t.co/cliCfhBZxN

[Chroma @trychroma]

Chroma supports multiple lexical search strategies for keyword-style retrieval.

FTS. BM25. SPLADE.

Walks through how they differ, and when each one wins.

[Chroma @trychroma]

Docs on SPLADE:

https://t.co/AzD6MkhBCg