nvidia/DeepSeek-V3.1-NVFP4
Different architecture vs deepseek-ai/DeepSeek-V3.1. Different architecture: only 1395 of 91991/180547 tensors match by name and shape.
Different architecture vs deepseek-ai/DeepSeek-V3.1. Different architecture: only 1395 of 91991/180547 tensors match by name and shape.