rapid-mlx/Qwen3.8-27B-4bit-MTP-MLX

Quantization vs Qwen/Qwen3.8-27B. rapid-mlx/Qwen3.8-27B-4bit-MTP-MLX is a quantized version of the other (U32/BF16).

1 derivatives checked

• rapid-mlx/Qwen3.8-27B-4bit-MTP-fp16-MLX: Same weights, converted

19 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse