Manuel Schmid

@MindOfMatter sadly this doesn't work on browsers using dark mode as white is the hardcoded background color.  Can you please have a second look?

--always-high-vram or --always-gpu serve the same issue, except they will actually keep the model loaded in VRAM, right? Default is to put a model on the offload device, which is...

I can imagine that this will lead to issues when you don't have sufficient memory to load new models to the list and keep it in memory. Would you please...

Just tested again, if it works it works, but there are obvious downsides. This only works well with high available memory, as when too much memory is used, the OS...

@midnitedestiny in this PR, tab files https://github.com/lllyasviel/Fooocus/pull/2120/files

As Fooocus is a SDXL only GUI with specific adjustments,novelty functions and individual controlnet / adapters his would lead to weeks/months of effort if even possible to guarantee the same...

@kubilaykilinc yeah, i briefly forgot about that ^^ This is indeed possible, but 1.5 models can't be used as base model.

Hey @outsami-9527, can you please elaborate how this should work and what you would expect from such an implenentation? Thanks!

@LIQUIDMIND111 there currently are only these 2 mix options - `mix image prompt and upscale/vary` - `mix image prompt and inpaint` I don't know where you see `mix upscale/vary and...

I haven't been experiencing the same kind of issue with Fooocus. Can you confirm that generated images change after 100x using the same settings with the same seed? Can you...