[!IMPORTANT] This quantization was uploaded on 2026-08-22 and replaces a previous version. If you downloaded this model before this date, please re-download for the updated weights.

Qwen3.8-27B-oQ4e-fp16-mtp

Quantization details

• Model type: qwen3_5
• Bits: 4
• Group size: 64
• Format: MLX safetensors