>>25034
Obviuskly you will not be able to use the same model as mine

I don't know which model to recommend to you but pretty much all good stuff is here
https://huggingface.co/TheBloke

you can select some 7B model in 4 bit mode. which will likely fit into your GPU somewhat but it will run pretty fast juts on the cpu
or try GGUF model and split it between CPU and GPU

some experimenting is needed to find what works best

I think you could start from this one:
https://huggingface.co/TheBloke/Kuchiki-L2-7B-GGUF
select which version you like and juts tell that text generation ui to download it.