AtlasCloud/DeepSeek-V4-Flash-0731-FP8-DSpark
Different architecture vs deepseek-ai/DeepSeek-V4-Flash-0731. Different architecture: only 1615 of 72317/72271 tensors match by name and shape.
Different architecture vs deepseek-ai/DeepSeek-V4-Flash-0731. Different architecture: only 1615 of 72317/72271 tensors match by name and shape.