Anthropic released Message Threads as a public beta on its Claude platform to change how API requests handle conversation history. Previously, every request required resending the entire chat log. The new system only transmits data that has changed since the last message. Anthropic states this reduces the time to first token by approximately 35 percent for requests containing between 500,000 and one million tokens.
Coding agents, computer-use agents, and workflows processing large volumes of documents or images benefit the most from this efficiency. The thread remains encrypted in memory between requests and expires after five minutes of inactivity. Token usage does not change because Claude still reads the full conversation history to maintain context. The feature operates on the Claude platform and through AWS, Microsoft Foundry, and Google Cloud.
* Only delta data is transmitted between requests
* Context window size and total token costs remain unchanged
* Threads expire automatically after five minutes of inactivity
