elyza

/

ELYZA-japanese-Llama-2-13b

like 20

Hi,

I am trying to deploy this model on an A10.1 card using TGI, but I am getting a GPU memory error as TGI tries to convert "pytorch_model.bin" into SafeTensor files. Kindly help in resolving this issue by merging the SafeTensor files.
https://huggingface.co/elyza/ELYZA-japanese-Llama-2-13b/discussions/1