Creating and testing the latest version of
Magical woman - V6 F32 | Stable Diffusion Checkpoint | Civitai
i rebuilt my workflow. I added CLIP SDXL support, which significantly improved the quality of the generated images.
Wf is mostly used to generate portraits and therefore includes a randomiser for faces and hairstyles:
Here u have my the best settings :) but if u will create better, pls write it in comments :)
Support Paweł Tomczuk on Ko-fi! ❤️. ko-fi.com/aderek514 - Ko-fi ❤️ Where creators get support from fans through donations, memberships, shop sales and more! The original 'Buy Me a Coffee' Page.
Visit my DeviantArt page: Aderek - Hobbyist, Digital Artist | DeviantArt
Description
FAQ
Comments (3)
Looks interesting! The .json file in the .zip archive doesn't contain a workflow, though. It has git_custom_nodes, file_custom_nodes, and pips. It doesn't contain any nodes.
i always put snapshot, now i was add json normal file and few images inclued wf too. GD ;)
Thanks for the example with KSampler Cycle.
Why is this done: for KSampler Cycle, both positive and negative prompts are taken from the same ClipTextEncoder with negative text as "g" and "l" input (2024-06-22_ComfyUI_Prompt_00004.png)?