zai-org

/

GLM-4.7-Flash

like 1.53k

Follow

Z.ai 9.65k

We’ve published two quantised variants of GLM-4.7-Flash for the community, optimized for different NVIDIA GPU generations:

FP8 (Hopper-class GPUs)
👉 https://huggingface.co/Geodd/GLM-4.7-Flash-FP8

W8A16 (Ampere-class GPUs)
👉 https://huggingface.co/Geodd/GLM-4.7-Flash-W8A16