primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8

Quantization vs Qwen/Qwen3.8-27B. primitive-ai/Qwen3.8-27B-mixed-NVFP4-FP8 is a quantized version of the other (U8/BF16).