Inferact/Qwen3.8-Flash-Next-NVFP4

Quantization vs Qwen/Qwen3.8-Flash-Next. Inferact/Qwen3.8-Flash-Next-NVFP4 is a quantized version of the other (BF16/U8).

2 derivatives checked

• provsalt/Qwen3.8-Flash-Next-NVFP4-PLE-NVFP4: Same weights, repacked
• codegyver/Qwen3.8-Flash-Next-NVFP4-FP8-PBWO-DGX-Spark: LoRA-style

On the Hub · Downloads on Model Pulse