INCModel/Qwen3-32B-MXFP8-FP8ATTN-CT-AutoRound

Quantization vs Qwen/Qwen3-32B. INCModel/Qwen3-32B-MXFP8-FP8ATTN-CT-AutoRound is a quantized version of the other (F8_E4M3/BF16).

On the Hub · Downloads on Model Pulse