Li Tan

I ran into the same problem. It seems that the guide was deprecated along with the release of Vertx 4. If you still want to check out the doc, you...

@samee , thanks for your reply! I worked around this problem, and I was able to build it successfully on a ubuntu server. I tried suggestions in your referred issues,...

Thank you, @simon-mo, @DarkLight1337 and @Tostino ! Is the PR ready to merge and incorporate into the upcoming release?

+1 The inference speed is very impressive. Amazing work! It will be awesome to run quantized model with it.

> it seems that MLX will speed this up greatly. I am not sure about this and look for more data points. I tested llama.cpp and mlx on the 16...

That is pretty interesting. @robertritz, do you have data points of your examples on the evaluation latency? Is your prompt templated? If yes, have you tried tokenizing some of them...

BTW, it seems that it can be easily fixed with ``` def load_demo(): - dropdown_update = gr.Dropdown.update(value=list(category_selector_map.keys())[0]) + dropdown_update = gr.Dropdown( + value=list(category_selector_map.keys())[0], interactive=True + ) return dropdown_update, dropdown_update ```

Thanks for your reply, @xiyang-aads-lilly ! In the case that we need to fine-tune on a small set of documents (

Thank you, @wmx-github ! It works now. Should we update the setup.py (https://github.com/QwenLM/Qwen-Agent/blob/main/setup.py) to specify the versions?

@tridao thanks for the information! What's our targeted timeline for supporting soft capping in the backward pass?