kinoc

Would like an optional encoding flag, which defaults to "utf-8" but you could specify others. I have to use "latin-1" for some cases.

I experienced problem when doing a txt2img generation, then stopping server to pull update of code, then try regeneration with the prior txt2img browser page. It appears the state associated...

Is the PR still active or planned?

Almost exactly the same. I am using ggml-vicuna-13b-4bit.bin and ```python server.py --auto-devices --listen --no-stream --verbose``` Also under windows but using an 3080. Similarly, the first request will run and return....