mlx-community/Qwen3.8-27B-MTP-mxfp8

Quantization vs Qwen/Qwen3.8-27B. mlx-community/Qwen3.8-27B-MTP-mxfp8 is a quantized version of the other (U32/U8).

1 derivatives checked

• t0rr3sp3dr0/Qwen3.8-27B-MLX-MTP-mxfp8: Partial fine-tune (fewer layers)

On the Hub · Downloads on Model Pulse