其他

mtplx

youssofal/mtplx

3x decode TPS increase On Qwen 3.6 27B @ temp 0.6 | Native MTP Speculative Decoding On Apple Silicon With No External Drafter.

去 GitHub 看仓库

标签

  • apple-silicon
  • local-ai
  • mlx
  • mtp
  • openai-compatible
  • qwen
  • speculative-decoding
  • anthropic-compatible

同类项目