Claude API message threads at a glance
In public beta, message threads keep a Messages API conversation on Anthropic's servers between turns, so each later request carries only new messages. Anthropic reports about 35% lower median time to first token on 500k to 1M token requests. Token usage and prices stay the same.
What changed with Claude API message threads
On November 2, 2026, Anthropic released message threads in public beta for the Messages API, behind the message-threads-2026-08-12 beta header. The first request sends the full conversation and starts a thread; later requests send only the new messages and the ID of the previous response.
Anthropic says threads are held in memory between turns, encrypted and not written to disk. A thread expires 5 minutes after its last request, after which the client resends the full conversation, and it is deleted no more than 1 hour after its last use. Threads work on the Claude API, Claude Platform on AWS, Claude in Microsoft Foundry and Claude Models on Google Cloud (Anthropic managed).
The gain is latency, not cost. Anthropic reports that median time to first token drops about 35% for requests of 500k to 1M tokens, measured where threads are on by default, which it says covers most Claude Code users. Claude still reads the full conversation every turn, so token usage, standard prices and rate limits are unchanged. These are vendor-reported figures.
Claude API message threads in numbers
Values are copied from the linked source on the date shown. They are not IA Compa measurements.
- Median time-to-first-token reduction (500k–1M token requests)as of · Source for Median time-to-first-token reduction (500k–1M token requests): claude.com (opens in a new tab)
- ~35% (vendor-reported)
Why Claude API message threads matters for businesses
For teams running long tool-calling agents, such as support bots or back-office automations with big contexts, this means smaller payloads and faster first replies. Keep your full-history fallback: any pause longer than 5 minutes ends the thread, and the feature is still in beta.
Sources for this story
- Make long-running agents faster with message threads (opens in a new tab)Official
Anthropic · claude.com
- Claude Platform release notes (November 2, 2026) (opens in a new tab)Official
Anthropic · platform.claude.com
- Message threads (opens in a new tab)Official
Anthropic · platform.claude.com
Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.
- #anthropic
- #claude-api
- #agents
- #latency
- #beta