↓ Skip to main content

mlx-realtime

Local realtime voice service for Apple Silicon. It speaks a subset of the OpenAI Realtime (GA) WebSocket protocol and runs every model on the Mac: Silero VAD and Smart Turn, then Qwen3-ASR, Qwen3.6 and Qwen3-TTS on MLX.

Repository: rasonyang/mlx-realtime

A Realtime Voice Bot on One Mac, and Why the Third Call Breaks It

A Realtime Voice Bot on One Mac, and Why the Third Call Breaks It

AICC answers calls with anything that speaks the OpenAI Realtime protocol. In the first post in this series that meant a hosted model, or Cascade Realtime Gateway, which runs ASR → LLM → TTS behind the protocol but still calls hosted providers for each stage. The question this post answers: can one Mac run the whole cascade itself, every model local, and still answer inside the deadlines AICC’s client enforces?