q5sys (JT)

> How do we change the setuptools version required in the mamba pulls? I couldn't find a list of dependencies anywhere in the easy diffusion code. You didn't find it...

You can try the patched bootstrap file I have in my fork: https://github.com/q5sys/easydiffusion/blob/main/scripts/bootstrap.sh Hopefully that will work for you all.

> ok that bootstrap.sh was not yet included in your fork if the git was cloned. > > To fix it I copied the code from your linked bootstrap.sh into...

I think this is a CUDA issue, which you wont be able to get around. A CUDA env expects to have full access to the device, it's not like a...

can you confirm the same thing I reported here: https://github.com/easydiffusion/easydiffusion/issues/1760 Where the overall speed gets split between two cards for no acceleration?

@Xeraster You can span an LLM across multiple GPUs because of how they work. LLMs function in a completely different way than the way stable diffusion does. Due to the...

I've run into something similar. I can train a job with everything working fine. I shut down my machine, start it up the next day (with no updates at all),...

It would be really nice if this information would be added to the documentation by someone on the team, or a community member who is active on discord. I realize...

That's just the manual way to use loras in a prompt. I updated the wiki with a small explanation. https://github.com/easydiffusion/easydiffusion/wiki/LoRA#manually-using-loras

I just ran into this. FYI, you can also install `python-sentencepiece` from the AUR and then when you run `pip install -r requirements.txt` it will pull from what you installed...