相关标签
macosinference-servermlxapple-siliconopenai-apillmswiftappleanecoreml

Here are 1,847 public repositories matching this topic...

Run Claude Code 100% on-device with local AI on Apple Silicon. MLX-native Anthropic-API server, 65 tok/s Qwen 3.5 122B, Llama 3.3 70B, Gemma 4 31B. Private, offline, airgap-ready. Built for NDA / legal / healthcare workflows.

  • Updated Jun 5, 2026
  • Python