Catalpa
Catalpa
> cool, getting no results yet, what do you enter in the Searx instance address? You need to host one yourself
> Why is everything called Searx? Is that the name name of the module? Even in the UI and code comments it's called "Searx" but now it is renamed to...
also fixes https://github.com/oobabooga/text-generation-webui/issues/629
> @catalpaaa Thx for the effort, just a kindly reminder: > > the theme.css missing is not the only problem, there is another problem that the real predict post request...
> How to test this? do a reverse proxy with nginx with some subpath, then boot with the same --subpath heres my nginx config: ``` location /textgen/ { proxy_pass http://localhost:1234/;...
> How to test this? i can show you a on/off demo on my personal server if you have a way to share it privately
> Is FastAPI guaranteed to be installed? for me, it came with gradio you can put it in req tho
> Is FastAPI guaranteed to be installed? fastapi is a dependence of gradio, it should be fine
> It was a bug of gradio, and it is fixed a few days ago by the gradio teams. > > > > just add your subpath into the _root_path_...
first of all model type should be llama 30b 4bit need way more spaces then 13b, try to use `--pre_layer PRE_LAYER` to offload