Trending repositories: qwen3-next

1 of 1 repository

• youssofal/MTPLX

  3x faster speeds on MLX | Qwen 3.8 27B | Native MTP Speculative Decoding On Apple Silicon With No External Drafter.

  AI summary: A native Mac app and CLI tool for running local language models significantly faster utilizing multi-token prediction.

  2,216ai-mlPythonApache-2.0anthropic-compatibleapple-siliconinference-enginelocal-ai