Adil3tr
Adil3tr
I've had the same issue. I was making images with LMS at 20 CFG and 50 steps and having no issues for weeks, but today I make ones at 12...
> The extended sampler settings using Karras in the UI are best approximations of good default parameters. If you'd like to experiment with https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Custom-Scripts#alternate-noise-schedules you may find better parameters. This...
> File names are so subject to change, though. The name of an image hardly represents the image itself. Could there be a solution that involves computing hashes for every...
If it's not fixable, they should maybe remove LMS from Automatic, because it has really serious artifacting most of the time right now.
I regularly run XY grids with dozens of checkpoints, it only takes a few seconds for most of them to load into RAM. If the prompts were calculated beforehand, it...
Yeah I guess the alternative would be adding the ability for "__token__" to always be replaced with the token of the model currently loaded, with changing the model left to...
> I would second something like this if it were possible, though with a way to specify a model, not just full randomize. maybe by putting a specific character/symbol on...
Just had this issue today, turning on the "template in PNG info" setting breaks PNG info's send to txt2img function.
> I definitely think this is doable and would work. It would still be an automatic calibration, and would allow any HMD to be used, regardless of drifting reference frames....