z_

 我看这个逻辑，似乎是只有gpt-4，名字里带turbo的，还有是writely service的，是走chat接口。

考虑到ChatGPT的回答往往很长，为追求回答的质量，我把问题生成和结果生成拆分开进行。

感谢贡献~ PyLint 报错，需要修改下。 SM2的接口感觉还是保持原样比较好，主要是出于兼容性的考虑，目前已经有人在用这个库了，升级之后需要不能break。所以也需要修改下。

https://github.com/deepseek-ai/DeepSeek-V3/blob/2f7b80eecebf3d1c84da5a0d465f6639ea175012/inference/fp8_cast_bf16.py#L80 I tried changing ```python new_state_dict[weight_name] = weight_dequant(weight, scale_inv) ``` to ```python new_state_dict[weight_name] = weight_dequant(weight.float(), scale_inv) ``` and then the conversion code seems to work now on A100.

> > [DeepSeek-V3/inference/fp8_cast_bf16.py](https://github.com/deepseek-ai/DeepSeek-V3/blob/2f7b80eecebf3d1c84da5a0d465f6639ea175012/inference/fp8_cast_bf16.py#L80) > > Line 80 in [2f7b80e](/deepseek-ai/DeepSeek-V3/commit/2f7b80eecebf3d1c84da5a0d465f6639ea175012) > > new_state_dict[weight_name] = weight_dequant(weight, scale_inv) > > I tried changing > > new_state_dict[weight_name] = weight_dequant(weight, scale_inv) > > to >...