Anthropic launches Claude Message Threads: ~35% lower time-to-first-token and smaller API payloads, no security impact
Matéria consolidada. Reúne 2 matérias que o Tifon tinha publicado separadamente sobre a mesma história. A IA reescreveu o relatório com o que as 2 fontes trouxeram em 05/11/2026 00:06; os links antigos trazem para cá.
Fontes: securityonline.info ↗ zdnet.com ↗
Anthropic launched Claude Message Threads, a feature of its Messages API (also used in Claude Code) that stops clients resending the full conversation history on every turn. Applications send only the new message plus the previous response identifier, while the server persists conversation state. Anthropic reports about 35% better time-to-first-token and a median payload drop from roughly 800KB to 1.8KB in long dialogues (securityonline.info). Both sources describe this as a product and performance announcement, not a security incident: no vulnerabilities, threats or security risks are discussed. Billing, token costs and rate limits are unchanged because the model still processes the full context on every pass (zdnet.com, securityonline.info). A hands-on test estimated real-world savings of 2-4 minutes per four-hour coding session, mostly for large-context sessions of 500K-1M tokens, and found no improvement in model reasoning or need for context compaction (zdnet.com).
Detalhes técnicos
- A thread is created with an initial request containing the full dialogue history; later requests submit only the newest prompt and the prior message ID (securityonline.info).
- Threads expire automatically after five minutes of inactivity. Referencing an expired thread returns an error, so clients must keep local transcripts to rebuild sessions (securityonline.info).
- Only sequential, append-only operations are supported. System prompts and tool schemas must stay constant for the thread's lifetime; changing them requires a new thread (securityonline.info).
- The feature is designed to complement prompt caching, combining lower network overhead with discounted cache-read compute costs for eligible context segments (securityonline.info).
- Defenders should note that conversation state is now persisted server-side, so data-retention and handling reviews may apply; the sources do not discuss this.
Linha do tempo
-
Anthropic Introduces Claude Message Threads to Reduce API Payload Size and Latency
ler a análise desta fonte ocultar
Anthropic has launched Claude Message Threads, a new feature in its Messages API that reduces conversational payload size and latency for multi-turn and agentic AI workflows. Instead of resending entire dialogue histories with each request, applications now send only incremental messages plus the previous response identifier, with the server persisting conversation state. Anthropic reports this cuts median payload size from roughly 800KB to 1.8KB in long dialogues and reduces time-to-first-token latency by about 35%. The change does not alter token-based billing, rate limits, or pricing, since the underlying model still processes the full context on each pass. This is a product/performance announcement with no security vulnerability or incident implications.
Technical Details
Threads are created with an initial request containing the full dialogue history; subsequent requests submit only the newest prompt and the prior message ID. Threads automatically expire after five minutes of inactivity, after which referencing an expired thread returns an error, requiring clients to retain local transcripts to reconstruct sessions. Threads support only sequential append-only operations, and core parameters like system prompts and tool schemas must remain constant for the thread's lifetime; any changes require a new thread. The feature is designed to complement prompt caching, combining reduced network overhead with discounted cache-read compute costs for eligible context segments.
-
Anthropic's Claude Code Message Threads Improve Latency, Not Costs or Security
ler a análise desta fonte ocultar
This is a product performance analysis, not a security incident. The author tested Anthropic's new message threads feature in Claude Code, which reduces repeated re-uploading of full conversation history on each API turn, improving time-to-first-token by up to 35% according to Anthropic. The author's own log analysis estimates 2-4 minutes of real-world savings per four-hour coding session, mainly benefiting large-context sessions (500K-1M tokens). The feature does not reduce token costs or rate limits, since the model still processes the full conversation each turn, and it does not improve model reasoning or reduce the need for context compaction. No vulnerabilities, threats, or security risks are discussed in this article.