-
I have this working with several models which all work well with a considerable increase in speed, I cannot get any SDXL models to work though.
However, if I add a negative prompt, and I've tried s…
-
I downloaded all the files for Fooocus but whenever I try to input an image for a variation or image prompt, I get this error:
`torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 30…
-
I have problem with "Inpaint":
> Error while Generating.
> The size of tensor a (53) must match the size of tensor b (54) at non-singleton dimension 3 (controlnet.py:769)
And also when using "M…
-
### Checklist
- [X] The issue exists after disabling all extensions
- [X] The issue exists on a clean installation of webui
- [ ] The issue is caused by an extension, but I believe it is caused by a …
CCpt5 updated
9 months ago
-
Hello. The NeCA model has been trained separately for each device type, with distinct generator and denoiser models for each. However, there is no information about device types in the SIDD and DND be…
-
### Describe the bug
Hi there, thanks for all the work.
After auto1111 update which added refiner and hiresfix to be able to be used at the same time on dev (https://github.com/AUTOMATIC1111/sta…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues and checked the recent builds/commits
### What would your feature do ?
After updating Controlnet 1.1.431, IP…
-
Not so many new features but I'm running it right now and the stability alone is worth it.
6.1 crashed a lot for me but 6.2.1 hasn't crashed a single time yet.
Other than stability I guess there s…
-
its kinda hard to explain so ill just send images
basically everything ends up super…
-
As stated. It is just not there. I got two other options though - ALBEDO and NORMAL. I have no idea what to do with this.