Joe Sandusky

Did training on 5b_lyrics model ever work out for anyone?

A user in one of the SD groups mentioned this new PR for python: https://github.com/pytorch/pytorch/pull/106200 Perhaps would solve the VRAM problem? ---- CUDA Unified Memory This PR adds support for...

Ok, so now new NVIDIA driver's have an option in control panel to allow to "Prefer Sysmem offload" which will use RAM in the event VRAM required memory is unavailable...

The above post has the solution: 1) Right click on the border of the cmd/console window 2) Under the Options tab/Edit Options uncheck "Quick Edit Mode" 3) Click OK/Close That...

Almost positive that's what he's talking about. Bugged me too until I asked ChatGPT 4 about it. Hopefully they'll reply and confirm you provided the solution they were seeking. Not...

There's a "temporary" extension for A1111 that will allow you to generate w/ Stable Cascade (it an independent tab) here: https://github.com/blue-pen5805/sdweb-easy-stablecascade-diffusers Hopefully it's natively implemented sometime in the near future.

I am having this issue now as well. I have a lot of wildcard text files where the subject is a variable ${nme}. I don't care about using the variable,...

Post an example image from a model you trained From your txt description it sounds like you may be using the wrong VAE for the model type (SDXL/v1.5). Glitchy overexposed...

I have this problem also but haven't narrowed down what causes it for me. Only that it's happening enough to be frustrating.

I've experienced the second error, "'NoneType' object is not iterable", when using an extension called Model Mixer (merges models). I found a workaround is to hit the button to refresh...