{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreibfnhqbkfz63gfwbjebgs6f3z5zo5jp6kqbahzpe3jc4p3efymwca",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mernfsengw32"
  },
  "path": "/t/m-courtyard-gui-app-for-fine-tuning-mlx-community-models-on-mac-open-source/173454#post_1",
  "publishedAt": "2026-02-13T12:32:49.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "GitHub - tuwenbo0120/m-courtyard: M-Courtyard: Local AI Model Fine-tuning Assistant for Apple Silicon. Zero-code, zero-cloud, privacy-first desktop app powered by Tauri + React + mlx-lm."
  ],
  "textContent": "Hi everyone!\n\nI built an open-source macOS app called M-Courtyard that provides a full GUI\nfor fine-tuning LLMs using mlx-lm on Apple Silicon.\n\nIt integrates directly with Hugging Face’s mlx-community models — you can\nbrowse, download, and fine-tune models like Qwen 3, DeepSeek R1, Llama 3\nand more, all through a visual interface.\n\nWorkflow:\n\n  1. Import documents → Auto-generate training data\n  2. Select a model from mlx-community (auto-download)\n  3. Fine-tune with LoRA/DoRA — real-time loss visualization\n  4. Test with built-in chat\n  5. Export to Ollama\n\n\n\nNo CLI, no Python scripts needed.\n\nGitHub: GitHub - tuwenbo0120/m-courtyard: M-Courtyard: Local AI Model Fine-tuning Assistant for Apple Silicon. Zero-code, zero-cloud, privacy-first desktop app powered by Tauri + React + mlx-lm.\nRequirements: macOS 14+, Apple Silicon, 16GB+ RAM recommended\n\nWould love to hear your thoughts, especially from fellow MLX users!",
  "title": "M-Courtyard: GUI app for fine-tuning mlx-community models on Mac (open source)"
}