spb/zyquo-mlx

Public MIT

The local MLX foundry for your Mac — run, fine-tune, quantize, and ship models. Nothing leaves your machine.

macos swift ai mlx fine-tuning

14commits
1branches
1releases
1009.0 KBon disk
maindefault branch
3 mo agolast push

Thu, Jul 30, 2026 4
1. phase8: Developer ID signing + notarization — spctl 'Notarized Developer ID' on first submission, stapled app + DMG, minimal entitlements; Definition of Done review complete
  …
  Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
  simon-pierre boucher 3 mo ago · 1 file +55 −4
2. phase2: architecture + inference POC — domain models, InferenceEngine actor, ModelStore, MemoryAdvisor, CLI POC (LLM 603 tok/s + embeddings verified)
  …
  Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
  simon-pierre boucher 3 mo ago · 1 file +1 −1
3. phase1: checkpoint — swiftbuild backend resolved (metallib OK), Makefile/BUILD.md updated, app assembled and launch-verified
  …
  Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
  simon-pierre boucher 3 mo ago · 1 file +7 −4
4. phase1: SPM manifest (pinned MLX deps), app entry + Apple-Silicon gate, Info.plist, Makefile, README
  …
  Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
  simon-pierre boucher 3 mo ago · 1 file +75