Block unveils Dessa, a full-duplex speech model trained only on speech

Block's AI research lab released Dessa, a speech-to-speech model trained only on human conversation audio, with no text and no pretrained models. It runs locally on a MacBook Pro M4 Max faster than real time and was trained on a single NVIDIA GB200 NVL72 rack (72 GPUs). No benchmarks, weights or API have been published yet.

• Dessa listens and talks at once, capturing laughter, sighs and interruptions
• Trained on one GB200 NVL72 rack versus 1,016 H100s for Moshi
• Runs locally on a MacBook Pro M4 Max faster than real time
• No published benchmarks, weights, paper or API