feat: add MLX 4-bit quantized model (Apple Silicon optimized)

feat: add DPO v6 GGUF Q4_K_M model

feat: replace SFT-only GGUF with DPO v4 merged Q4_K_M

feat: add Q8_0 GGUF quantized model (7.6GB)

feat: add Q4_K_M GGUF quantized model (4.4GB)

Upload folder using huggingface_hub

initial commit