Spaces:

artificialguybr

/

qwen-14b-chat-demo

like 81

Sleeping

Also, I copy the files in this space to local, and run the app.py, but there is no output shown in the text box, but there are outputs at the termnial where the code is run. Please help, I want to run this on my local machine. But need to increase the maximum output tokens and get the output to the textbox.

I don't plan to increase output tokens. It's a necessary choice to avoid long queues.

About running local: I can't help. I have no experience of running locally. I only run things here at HF Spaces. I don't have a GPU in my notebook.

Thanks for the reply, I want to change the maxium I find somewhere that you need to change the:
"max_new_tokens": 512,
in the model's
generation_config.json

Do you know if that's the case, or there is somewhere you have set it in the app.py?

I think the problem might be the box for reponse is too small, in below, the first is the working one (huggingface), the second is the none working one (local), not sure how to change that.

Find the problem: Gradio version too high. Solution: pip install gradio==3.41.0