All versions require PickScore nodes to be installed.
Gemini 2.0 version also requires Gemini 2.0 Flash to be installed.
All versions that are using Gemini 2.0 are updated and are also using system prompt node that can be found in the creepy_nodes pack on github and in ComfyUI manage.
NOTE: If you have an issue loading the Gemini node, try to pip install sounddevice
This is a fully automatic workflow, mix the Detailer Deamon and Flux Redux, as well as Flux Dev and Stable Diffusion 3.5 Turbo.
I have included 2 nodes to clear VRAM/Cache at strategic places, but I can't say what the minimum system requirements might be. I have 12GB VRAM and 32GM system RAM, and the workflow is working well for me.
1 generation creates 5 images, of which one is upscaled, ran through detailer and finished with Stable Diffusion 3.5 Turbo.
If you have missing nodes, just go in tho manager and install them from there.
Upload an image and Florence will extract a prompt from it. Redux will make sure the images keep the style/theme and Detail Deamon will produce 4 images with variation on every run.
The images are then evaluated by Scorepicker and the best of the 4 images are choosen (the other 3 are also saved).
The choosen image will go through an upscaler. And after that the workflow changes the checkpoint from Flux dev to Stable diffusion 3.5 Turbo. With the SD 3.5 checkpoint the final image then go through a face detailer and finally through Ksampler with low denoise (0.15-0.3).
Description
Utilizing both Flux and SD 3.5 in the same workflow.
FAQ
Comments (7)
The Problem is that the end result turns up cartoony,the Image at the Upscale + Dump VRAM (Image B) looks better,is that normal or did you mixed some nodes accidently?
Are you using version 1.1? It has happen to me a few times at random, but not consitant. You can change the scheduler in the Ksampler from ddim_uniform to something else, and lower the denoise to between 0.1-0.2. That should fix it.
Did you remove the workflow? Because in the File its only Photo I got. Nothing else is there.
The workflow is in the image. Just drag it to comfy in the web browser.
Hey, thanks for sharing your workflow. Is there a way to include a regular positive prompt instead of using Florence? I haven't been getting the best results with it.
Thanks! Yeah that should be possible. I have been thikning of making a version like that actually, because even though Florence usually do a good job, sometimes I want to add or remove something from the generated prompt.
I'll see if I can upload a workflow without it in a couple of hours.
I've upoaded workflow with manual prompt now.