Thinking Machines Lab announces interaction models that process input and generate responses simultaneously, mimicking natural conversation

Thinking Machines Lab, founded last year by former OpenAI CTO Mira Murati, announced interaction models on Monday that enable AI to listen and respond at the same time rather than in sequential turns.

The company's model, TML-Interaction-Small, uses full duplex technology and responds in 0.40 seconds, claimed to match natural human conversation speed and be significantly faster than comparable models from OpenAI and Google.

A limited research preview is coming in the next few months, with a wider release planned for later this year; the current announcement is a research preview, not a public product release.