Anthropic Message Threads Cut Claude Agent Wait Times by 35%
Anthropic has launched message threads in public beta, letting Claude agents skip resending whole chats. On very long sessions, the wait before replies drops about 35%.
Anthropic has launched message threads in public beta for its API, a feature that lets developers building AI agents on Claude stop resending the entire conversation with every request. The company says it cuts the typical wait before Claude starts replying by about 35% on very long sessions.
How threads work
A coding agent can make hundreds of requests in a single session, and each one normally carries every earlier message and tool result. Once that history reaches hundreds of thousands of tokens, resending it adds a noticeable delay before Claude begins to answer.
With threads, an agent sends the full conversation once. After that, it sends only the new content, plus the ID of Claude's last reply.
Anthropic has already enabled threads for most users of Claude Code, its coding agent. On requests of 500,000 to 1M tokens, the company says the typical wait before Claude starts answering fell by about 35%.
Speed, not savings
The feature saves time, not money. Claude still reads the whole conversation on every turn, and Anthropic says prices and rate limits are unchanged. On cost, the company cut cache prices by 75% in September, as long agent sessions push up AI bills.
There are trade-offs. Threads 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. Threads are also not eligible for Anthropic's zero data retention or HIPAA health-data setups, which rules them out for some regulated workloads.
Availability
The beta works on all Claude models, both on Anthropic's own platform and through AWS, Microsoft Foundry and Google Cloud. It follows the release in September of Opus 5.5, a cheaper Claude model.
Apps worth a look
SUMO: the number puzzleTap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Dialo: learn languages by speakingTalk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
ediviewer: read EDI files like documentsOpen EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
More from this day
See all →
SuperMe raises $20 million to connect AI profiles of experts
BMW to Launch Level 2 Urban Driving Assist in Germany in 2027
Grok Cuts Chatbot Usage Limits as Musk Cites Unsustainable Costs
