Turns stylistic/anime/cartoon images into more realistic images. Definitely not realistic, but much more real than originally. The output style is very biased, I created the dataset myself so it was fairly limited.
I iterated through this working on the dataset a bunch of times and this one by far was the winner from all my other methods and attempts. It's incredibly hard to get flux to keep expressions, I was really trying to solve for that but failed. Maybe next iteration I'll try a higher weight.
Description
Best version from all of my attempts - it's far from perfect, but I've found it useful
FAQ
Comments (4)
This looks really good but could we have a workflow or a png with an integrated workflow? Pretty please?
It's literally the comfyui default workflow + lora loader. The reason i don't include one is because the workflow I use to generate these comparison images is a tangled mess and some custom nodes I made.
I can attach a workflow to this... but again, it's literally the default flux.2 klein 9b image edit template in comfyui with a lora loader added.
I'm currently training a lora, but when that ends I'll spin-up comfy and save off a workflow.
Added a workflow for you in the optional files
Thanks
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.