--- title: Anthropic adds message threads to the Claude Messages... description: Anthropic's Messages API now supports message threads in public beta, letting developers skip resending the full conversation on every turn. The server caches... canonical: https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Anthropic adds message threads to the Claude Messages API in public beta | daily.dev og:description: Anthropic's Messages API now supports message threads in public beta, letting developers skip resending the full conversation on every turn. The server caches... og:url: https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii og:image: https://api.daily.dev/og/posts/a4vFRfriI.png og:image:alt: Anthropic adds message threads to the Claude Messages API in public beta og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Anthropic adds message threads to the Claude Messages API in public beta **[Collections](https://daily.dev/sources/collections)** · 2 min read · 1 upvotes · 0 comments ## Summary Anthropic's Messages API now supports message threads in public beta, letting developers skip resending the full conversation on every turn. The server caches the conversation between turns, so you send only new messages plus the previous message ID. On requests of 500,000 to 1M tokens, time to first response drops by about 35%. It doesn't reduce cost: pricing and rate limits stay the same. Threads live only in encrypted memory and expire five minutes after the last request, so you still need your own copy of the conversation. The beta is available on Anthropic's platform, AWS, Microsoft Foundry and Google Cloud, and is already on for most Claude Code users. ## Content Anthropic has put message threads into public beta on the Messages API. The feature targets a specific annoyance: in a long agent session, every turn used to mean resending the entire conversation. ## How it works The API now keeps your conversation on the server between turns. You send the full conversation once. After that, each request carries only the new messages plus the ID of the previous response, and Claude starts replying sooner. ## What it changes, and what it doesn't On Claude Code requests between 500,000 and 1 million tokens, median time to first token dropped by about 35%. That's a speed gain, not a savings. Claude still processes the whole conversation on every turn, so token usage, prices and rate limits are unchanged. Only the amount of data you transmit shrinks. ## Expiry and storage Threads expire five minutes after the last request. Active sessions refresh the timer, and SDK helpers can resend the history automatically if a thread lapses. Threads are held in memory, encrypted and never written to disk. That means developers should keep their own copy of the conversation. ## Availability The beta runs on Anthropic's platform, AWS, Microsoft Foundry and Google Cloud. It is already on for most Claude Code users. ## Questions this post answers ### What does the message threads feature in the Claude Messages API do? It lets you send only new messages plus the ID of the previous message instead of resending the full conversation, since the server caches the conversation between turns. This cuts the typical wait before Claude starts replying by about 35% on requests of 500,000 to 1M tokens. It does not reduce cost: prices and rate limits stay the same as before. _Building long-running Claude agents? Track API changes like this via daily.dev before they break your setup._ ### How long does Anthropic keep cached conversation threads in the Messages API beta? Cached threads expire five minutes after the last request. They are held only in memory, encrypted, and never written to disk, so developers must keep their own copy of the conversation since the server's cached version can disappear after the timeout. _Developers designing agent fallback logic can follow API nuances like this on daily.dev._ ### Which platforms support the Claude Messages API message threads beta? The beta works on Anthropic's own platform, AWS, Microsoft Foundry, and Google Cloud. It is already enabled by default for most Claude Code users. _Choosing a cloud provider for Claude integration is easier when you track rollout details on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 2 discussions and 204 comments across x (as of 2026-11-03). **TL;DR:** Developers broadly welcome server-side message threads as a fix for the long-standing pain of resending full conversation history each turn, cutting payload size and latency, though many clarify it doesn't reduce token usage or cost since the model still processes the full context. **Sentiment:** 45% positive · 35% mixed · 20% skeptical **The case for** - Big latency win for long-running agent sessions since the client no longer resends massive conversation histories each turn. - Removes the need for developers to build custom state-management layers to track context. - Already enabled by default for most Claude Code users, including Pro/Max subscribers. **The pushback** - Clarified repeatedly that token usage, costs, and rate limits stay unchanged since the model still reprocesses the full conversation. - Server-side storage raises data-retention and compliance concerns, especially for zero-data-retention or regulated orgs. - Five-minute thread expiry is seen as too short for workflows with pauses like human review or long test runs. - Unclear how branching, editing earlier turns, or retries interact with the cached thread/message-ID chain. - Not supported on Amazon Bedrock, limiting availability for some enterprise users. **By community** - x (mixed): Reactions mix enthusiasm about reduced latency and payload size with pointed questions about cost, retention, expiry, and branching behavior. **Hottest debate:** Whether this change meaningfully reduces cost/token usage or is purely a latency and payload-size improvement. **Open questions** - Does cached thread state bill at the cheaper cache-read rate or full price? - Is the five-minute expiry configurable for longer-running or paused agent workflows? - How does editing or branching an earlier message interact with the server-side cached thread state? - Will this become available on Amazon Bedrock or remain limited to the current platforms? **Highlights** > @ClaudeDevs Server caching changes the economics more than the latency. When you stop resending full history every turn, long agent workflows stop punishing you for memory. The question is pricing: cached state is only a win if it stays cheaper than the tokens it replaces. > — [vargasjmh on x](https://x.com/vargasjmh/status/2117338411207897491) > @ClaudeDevs everyone spent the last two years writing custom sqlite wrappers to juggle context arrays just for Anthropic to finally hold the state server side. > — [shadowaguy on x · 3 points](https://x.com/shadowaguy/status/2117336280778555819) > @ClaudeDevs Yeah nice, but it doesn't reduce tokens cost :( Claude still reads the entire conversation on every turn & token consumption remains unchanged > — [\_navicstein on x · 1 comments](https://x.com/_navicstein/status/2117398711491781011) > @ClaudeDevs Server-side thread state is a real latency win, but it quietly moves your cost model: cache writes bill at a premium over base input tokens, and once the TTL lapses mid-session you're paying full price to rebuild context. Budget for that, not just the faster first token. > — [fishjar on x](https://x.com/fishjar/status/2117348091582087607) > @ClaudeDevs If a thread branches, or an earlier message is edited, does the cached server state follow the new branch, or does it stay tied to the original message ID chain? > — [BMTestsAI on x · 1 points](https://x.com/BMTestsAI/status/2117336452480819268) **Source threads** - [x](https://x.com/ClaudeDevs/status/2117332449382510884) · 0 points · 204 comments - [x](https://x.com/iaenunminuto/status/2117379389985825041) · 0 points · 0 comments ## Similar posts on daily.dev - [Claude Code cache confusion as Anthropic tweaks defaults, but quotas still drain](https://daily.dev/posts/claude-code-cache-confusion-as-anthropic-tweaks-defaults-but-quotas-still-drain-uvn7zes3k) · DEVCLASS · 5 upvotes · 1 comments - [Claude devs complain about surprise usage limits](https://daily.dev/posts/claude-devs-complain-about-surprise-usage-limits-tbjb8dqml) · The Register · 0 upvotes · 0 comments - [Anthropic puts Claude inside Slack as a tagable co-worker](https://daily.dev/posts/anthropic-puts-claude-inside-slack-as-a-tagable-co-worker-9ncf1jwhw) · TechCentral · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#llm](https://daily.dev/tags/llm), [#architecture](https://daily.dev/tags/architecture), [#claude](https://daily.dev/tags/claude), [#anthropic](https://daily.dev/tags/anthropic) [View this post on daily.dev](https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Anthropic adds message threads to the Claude Messages API in public beta","url":"https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii"},"datePublished":"2026-11-02T20:18:09.857Z","dateModified":"2026-11-03T13:49:32.199Z","description":"Anthropic's Messages API now supports message threads in public beta, letting developers skip resending the full conversation on every turn. The server caches...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ef24317fe03464711d7871f80365157f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ef24317fe03464711d7871f80365157f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,llm,architecture,claude,anthropic","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Anthropic adds message threads to the Claude Messages API in public beta"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/anthropic-adds-message-threads-to-the-claude-messages-api-in-public-beta-a4vfrfrii#faq","mainEntity":[{"@type":"Question","name":"What does the message threads feature in the Claude Messages API do?","acceptedAnswer":{"@type":"Answer","text":"It lets you send only new messages plus the ID of the previous message instead of resending the full conversation, since the server caches the conversation between turns. This cuts the typical wait before Claude starts replying by about 35% on requests of 500,000 to 1M tokens. It does not reduce cost: prices and rate limits stay the same as before. Building long-running Claude agents? Track API changes like this via daily.dev before they break your setup."}},{"@type":"Question","name":"How long does Anthropic keep cached conversation threads in the Messages API beta?","acceptedAnswer":{"@type":"Answer","text":"Cached threads expire five minutes after the last request. They are held only in memory, encrypted, and never written to disk, so developers must keep their own copy of the conversation since the server's cached version can disappear after the timeout. Developers designing agent fallback logic can follow API nuances like this on daily.dev."}},{"@type":"Question","name":"Which platforms support the Claude Messages API message threads beta?","acceptedAnswer":{"@type":"Answer","text":"The beta works on Anthropic's own platform, AWS, Microsoft Foundry, and Google Cloud. It is already enabled by default for most Claude Code users. Choosing a cloud provider for Claude integration is easier when you track rollout details on daily.dev."}}]} ```