Florence2model -

Hi ! Not sure to be at the right place but .. 

I just dockerised comfyUi  -> docker desktop and wsl2 

For image génération everything is ok ! 
BUT 
For LLM they are extremly slow to load and to run ( they are loaded with CPU .. ) 

Do you know how to change this for the GPU ? I order to load the model quickly and run it faster ! 

Thanks 