Chroma Headlines

BotBeat

Chroma releases research on 'Context Rot' showing LLM performance degrades with longer input tokens, challenging common benchmarks.

Companies:Chroma

The New Stack

This article is a guide on how to build a private document search app with RAG and ChromaDB.

Companies:Chroma

Level Up Coding

The article discusses Chroma Context-1, a 20 billion parameter large language model built specifically for retrieval tasks, and how it improves RAG systems by being a self-editing search agent.

Companies:Chroma

Geeky Gadgets

The article introduces Chroma's Context-1, a 20 billion parameter agentic search model designed for retrieval-augmented generation (RAG), highlighting its precision, speed, and cost-efficiency compared to larger models like GPT-5.

Companies:Chroma

MarkTechPost

Chroma has released Context-1, a 20B parameter agentic search model designed to act as a specialized retrieval subagent. It aims to solve the "context rot" problem in LLM agents through self-editing context and offers significant efficiency gains over larger general-purpose models.

Companies:Chroma

Pixels and Pulse

This article analyzes Chroma's Context-1 self-editing feature, questioning its practical usability due to the lack of an agent harness and raising concerns about state consistency and potential data loss in long-running agentic processes. It suggests architectural recommendations for integrating such a system.

Companies:Chroma

Trend Micro

The article from Trend Micro discusses the "Fault Lines in the AI Ecosystem" as detailed in the TrendAI State of AI Security Report. It covers current security challenges and vulnerabilities within the artificial intelligence landscape, highlighting risks and potential threats associated with AI technologies. The report aims to provide insights into the state of AI security and the evolving threat environment.

Companies:Chroma