Thinking Machines Lab Claims Real-Time Processing of Audio, Video, and Text

Mira Murati's Thinking Machines Lab has unveiled AI models trained to handle audio, video, and text at the same time, without the processing delays that typically plague current systems. Most AI today processes these inputs one after another, creating lag. The new architecture, previewed May 11, aims to fix that bottleneck in conversational AI. Benchmarks remain undisclosed.