Thinking Machines Lab, founded by former OpenAI CTO Mira Murati, has announced a new class of AI called interaction models, designed to process input and generate responses simultaneously rather than sequentially. The approach, known as full-duplex communication, enables the AI to respond mid-conversation in a manner closer to a natural phone call than a turn-based text exchange.
The company’s initial model, TML-Interaction-Small, claims a response latency of 0.40 seconds — roughly matching natural human conversational speed and reportedly faster than comparable models from OpenAI and Google.
The announcement is currently a research preview rather than a public product launch. A limited research preview is expected within months, with a broader release planned for later in 2026. Real-world performance remains to be independently verified.
