Hugging Face's logo

THUDM
/
chatglm2-6b-int4

Transformers
PyTorch
Chinese
English
chatglm
glm
thudm
custom_code
Inference Endpoints
Model card Files Community
22

quantization_code 源码是否开源呢?

#7
by Buxian - opened
Discussion
Buxian

quantization.py 中 quantization_code 的原始.cu源码计划开源吗

zxgov

不能反编译吗

Papersnake

写成这样是因为 hf 的限制，原始代码可以看 https://github.com/THUDM/GLM-130B/blob/main/cuda/quantization.cu 和 https://github.com/prnake/haruka-cuda/tree/master/quant-gemm

Buxian

好的，感谢！

Buxian changed discussion status to closed

Sign up or log in to comment