First Test and feedback, Pls add the possibility to connect to an independant running LLama server URL

[ClaudeStabile]

Dele,,

This is my first test with this plugin.
Please : do not use that with DOCKER !!!
Console issue :
Console language issue : my console is in French

This is downloading a Hugging face model. Please do no do that on your docker container as it will be a killer.... app. We need to have a separate volume so large files are mounted outside the container.
This is CPU based so will not be terrific for performances

###Perfect world i dream about :

• What i Expect is just enterring the URL of a private server as https:chatgpt.free-solutions.ch Not download locally the model or keeping that as an additionnal option
• Please try out with my server https://chatgpt.free-solutions.ch If we have this we can plug openfire to any IA server.

FYI : I am also capable to build Llama full GPU docker containers & servers. For now i run this at home but if we have serious sponsors we can do a Llama servers in datacenter. Cost for server, i quote that 30-70kCHF/1U Box with 48-96GB GPU nvidia. I have the supplier

Thanks for doing this great JOB :)
I need to have the possiblity to enter my sever URL as https://chatgpt.free-solutions.ch or http://chatgpt.free-solutions.ch:8080

Congrats