intel/neural-speed

is qwen been supported?

Closed this issue · 1 comments

Hello, I noticed that the llmruntime in ITREX seems to be well migrated to this project, I tried to use model.init() to load the qwen HF model, but python throws the error TypeError: Unspported model type qwen!
In my impression, the ITREX project supports the use of the qwen model, but it seems that after the update, the qwen model can no longer be used. Is there any problem with the qwen model? Will the qwen model be supported again in the future?

Hi @kunger97 ,

Thank you for comment. We have resolved this issue. Please give it a try.