Anthropic has launched message threads, a beta feature that lets developers building AI agents on Claude send a full conversation once and then only new messages plus the ID of Claude's last reply. The feature cuts the delay before Claude responds on long sessions, but it does not change pricing or rate limits.

Anthropic has launched message threads, a beta feature that lets developers building AI agents on Claude stop resending entire conversations with every request, the company said Monday.

The problem is most acute with coding agents. A single session can involve hundreds of requests, and each one normally carries every earlier message and tool result. Once that history grows to hundreds of thousands of tokens, resending it all adds a noticeable delay before Claude begins to respond.

How do message threads work with Claude agents?

Threads change the pattern. The agent sends the full conversation once. After that, it sends only what is new, along with the ID of Claude's last reply.

Anthropic has already switched threads on for most users of Claude Code, its coding agent. On requests of 500,000 to 1 million tokens, the company says the typical wait before Claude starts answering fell by about 35 percent.

Do message threads reduce Claude API costs?

The feature saves time, not money. Claude still reads the whole conversation on every turn, and Anthropic says prices and rate limits have not changed. The company addressed cost separately in September, when it cut cache prices by 75 percent as long agent sessions pushed up AI bills.

How long are Claude message threads stored?

Developers should plan for how threads are stored. They are held in memory, encrypted and never written to disk. They expire five minutes after the last request, so developers must keep their own copy of each conversation. Anthropic also says threads are not eligible for its zero data retention or HIPAA health-data setups, which matters for companies in regulated industries.

The beta works on all Claude models. It is available on Anthropic's own platform and through AWS, Microsoft Foundry and Google Cloud. The launch follows Anthropic's cheaper Opus 5.5 model, which the company released in September.

This article was written with the assistance of AI.
News Factory APP - agentic news to boost your SEO & AEO.