Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
Resource directory
Search Sweet Tea Pipes, external workflows, models, node packs, and guides in one evidence-linked catalog.
Start with a workflow instead ↗Search results
20 shown of 28,887
flux-1-dev-non-commercial-license
Using the 'retro artstyle' tag will help you generate an art style that is closer to the original source.
Modelo de clasificación multiclase de lesiones cutáneas dermoscópicas entrenado sobre HAM10000 (ISIC 2018 Task 3) mediante fine-tuning progresivo. Segunda etapa del pipeline de análisis de lesiones cutáneas —...
timm DFN CLIP (image encoder only) weights from
Works best at 0.75 weight.
This is an embedding of a woman that I've generated consistent faces of thru SD and decided to make into an embedding for easier access in the future. It uses a mere 5 tokens.
Another "Nobody" embedding. I have been playing around with different hair and eye colors, for my first three. I was one curious about how they would actually impact the training. After this one I will be focusing on...
Style Lora in the Style of League of Legends Wild Rift. If the weight gets higher u get typical lol elements, like hair ornaments, more detailed extra thingies on clothes and lol like backgrounds.
✨ Welcome to my page! I really hope you enjoy using this model and have fun creating with it. It means a lot to see what others can bring to life with these LoRAs, so please don’t hesitate to share your images and...
This is the large variant of the PMT-DINOv3 model trained for panoptic segmentation on COCO at 1280x1280 resolution.
Orqo Hype is a specialist composition model for the HyperFrames stack. Given a creative brief, it produces a staged pipeline:
Another merger
采样器:DPM++ 2M Karras,DPM++ SDE Karras,DDIM
This is still very early and I barely even tested it, but from the results I've got they're pretty good as far as I can tell, I'll release it so I can gen a few on-site, hopefully y'all enjoy it. I'm still dipping...
Tracie doesn't exist. This embedding was trained entirely on prompts, an will reliably create cat ears, no need to prompt for them.
