Anthropic releases Message Threads public beta for Claude, cuts time-to-first-token ~35% on very long conversations
Anthropic released Message Threads as a public beta for its Claude platform. The feature transmits only new content since the last message (rather than resending the full conversation), which Anthropic says reduces time to first token by about 35% for requests between 500,000 and one million tokens; token usage remains the same, threads are encrypted in memory and expire after five minutes, and the feature is available on Claude as well as via AWS, Microsoft Foundry, and Google Cloud.