Claude now answers long chats about 35 percent faster

Claude can now start replying sooner in very long conversations. Anthropic released a feature called Message Threads that cuts the wait for the start of a reply by about 35 percent, according to the company. It is a public beta, which means it is open to use but still being tested.

What changed

Until now, software talking to Claude through the Messages API had to resend the whole conversation with every request. Message Threads sends only what has changed since the last message.

The 35 percent gain is for time to first token. That is the pause before the first piece of a reply appears. It applies to requests between 500,000 and one million tokens. A token is a small chunk of text, often a word or part of one.

Anthropic says coding agents, computer-use agents, and workflows that process lots of documents or images benefit the most. Agents are programs that carry out tasks in many steps.

What stays the same

• Token usage does not change, because Claude still reads the full conversation history.
• The saved thread stays encrypted in memory and expires after five minutes of inactivity.
• It works on the Claude platform and through AWS, Microsoft Foundry, and Google Cloud.

Several things are not yet clear from the announcement. It does not say how much shorter conversations gain, or when the beta will end. Details are in Anthropic's documentation.