Car5577/mlx-to-gguf-apple-silicon

mlx-to-gguf-apple-silicon Convert an MLX 4-bit fine-tuned model (the output of mlx_lm.fuse) into a standard safetensors model that llama.cpp can convert to GGUF — so you can run your MLX-trained model in Ollama on a Mac. This solves a long-standing gap for the Apple Silicon local-AI community: fine-tuning with MLX is the natural workflow on a Mac, but the resulting quantized model could not be converted to GGUF for Ollama. The existing solutions (e.g. Unsloth) require an NVIDIA… See the full description on the dataset page: https://huggingface.co/datasets/Car5577/mlx-to-gguf-apple-silicon.