Qwen
/

Qwen2-VL-72B-Instruct-GPTQ-Int8

Image-Text-to-Text

8-bit precision

Model card Files Files and versions Community

Qwen2-VL-72B-Instruct-GPTQ-Int8 / model-00018-of-00021.safetensors

Commit History

fix(pad zero) pad intermediate_size to 29696 to make sure quantized model can use 8 tensor-parallel in vllm

d1eab90

可亲 commited on Sep 24

Upload folder using huggingface_hub

8ca59d7
verified

clonefy commited on Sep 17