A pretty simple workflow with a few common features. There are some important (more or less) notes inside the workflow. It is designed for 'legacy' mode. For now, I still think that Nodes 2.0 are mostly buggy and slower garbage, even though they have a few cool things. The metadata of a final image will not contain the diffusion model name; I still haven't found a good way to add it to the metadata. Fixed in v.1.2
I didn't notice that Lora doesn't get into the metadata for automatic detection. Fixed in v2.0.
Description
V1.7:
Turned back the previous method of the mode selector.
Added SeedVR2 upscaler.
Made some internal changes (for example, changed the image resize node).
V1.9:
Added comparison of nodes
A few node repositions, and other minor changes.
v2.0:
Fixed my mistake when Lora doesn't get into metadata.
A few minor changes.
FAQ
Comments (5)
Workflow is fast, but img2img work very bad. At least you tried, well done!
What exactly is very bad?
@DemonAlone
Character consistency. I upload a photo and enter the prompt to change the outfit, but the face changes to something random
@Gavr ,try to use the "inpaint" toggle and a mask or inverted mask, like I wrote in the workflow note. It doesn't work really well, but I don't think this is exactly an issue with my workflow. With a few tries, it's possible to get fine results.
idk if this is the problem since I'm pretty new to this, but when I tried img-2-img for the first time(on some other workflow) it ran very slow and then I realized the speed of the generation is depedant of the pixel resolution of the image intially input-ed.