Claude API adds message threads in beta, so apps send only new messages each turn

On 2 November 2026 Anthropic opened message threads as a public beta on the Messages API. A thread keeps a conversation on Anthropic's servers for a short time, so after the first request an application sends only the new messages and the id of the previous response instead of the whole conversation. Requests get smaller, and the time to first token stays lower and steadier as a conversation grows, which helps most in long agent loops.

Developers turn it on with the message-threads-2026-08-12 beta header and a thread parameter: the first request creates the thread with the full conversation, system prompt and tools, and later requests continue it. The SDKs include a thread helper that builds these requests. Threads work on the Claude API, Claude Platform on AWS, Claude Models on Google Cloud (Anthropic managed) and Claude in Microsoft Foundry.

Threads are best-effort. A thread expires 5 minutes after its last request and is deleted no more than 1 hour after its last use. When a thread can no longer be continued, the API returns a 404 error and the application has to send the full conversation again, so Anthropic tells developers to keep their own copy. Message threads are not eligible for zero data retention or HIPAA.