AngelBottomless

@gel-crabs Now it should work for both models!

@gel-crabs Resnet level 0, which is max as supposed to be - VAE max tile size was set to 128, swap size 6 The logs are removed!

@gel-crabs HR fix will use 100% cache (if option enabled, and, actually success / failure rate is now requires rework, some are steps and some are function call...) But I...

https://github.com/Mikubill/sd-webui-controlnet/blob/main/scripts/hook.py#L425 Okay, this explains why we have bunch of more big code....

https://github.com/aria1th/sd-webui-controlnet/tree/maybe-deepcache-wont-work I was trying some various implementation including diffusers pipeline, and I guess it does not work well with ControlNet.... https://github.com/horseee/DeepCache/issues/4 ControlNet obviously handles timestep-dependent embedding, which changes the output...

@gel-crabs Yeah, most of the U-Net forward hijacking functions won't work with this, It assumes the nearby step's effects are similar. Some more academical stuff: DDIM works well with this....

@bigmover https://github.com/aria1th/sd-webui-deepcache-standalone Use the extension please, and note that it can't be used with controlnet / some other specific unet hijacking extensions

@w-e-w I don't really expect users to modify the callback execution order - most of the case, they would not know anything about how extensions were written and working... rather,...

@w-e-w That makes sense, I was just trying to handle functionality setups - to just allow devs to plan and calculate their order, but those type of processing has to...

At least I'll suggest ```py p = StableDiffusionProcessingTxt2Img(sd_model=shared.sd_model, **args) try: p.scripts = script_runner p.outpath_grids = opts.outdir_txt2img_grids p.outpath_samples = opts.outdir_txt2img_samples shared.state.begin() if selectable_scripts is not None: p.script_args = script_args processed =...