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
17 shown of 34,588
Video Generation on a Laptop
Figure 1: ColBERT's late interaction, efficiently scoring the fine-grained similarity between a queries and a passage.
Update Jun 5, 2026: ☁️ Deploy Chronos-2 on AWS with AutoGluon-Cloud. Real-time, serverless, or batch inference in 3 lines of code — pandas DataFrames in, forecasts out. Check out the new deployment guide.
Using this open-source model in production? Consider switching to pyannoteAI for better and faster options.
license: apache-2.0 datasets: sentence-transformers/msmarco language: en basemodel: cross-encoder/ms-marco-MiniLM-L12-v2 pipelinetag: text-ranking libraryname: sentence-transformers tags: transformers
A MobileNet-v3 image classification model. Trained on ImageNet-1k in timm using recipe template described below.
ELECTRA is a new method for self-supervised language representation learning. It can be used to pre-train transformer networks using relatively little compute. ELECTRA models are trained to distinguish "real" input...
This updated version of the 4K Wallpaper Desktop Creator leverages the Flux1 or Flux1Dev base model, providing a refined workflow for image processing.
This is my basic workflow, which I developed after many learned. It's a balance between quality, speed, performance and facilities. Every new work I will start, I start with this.
This workflow is designed for creating IP-Adapter FaceID Portraits, compatible with both SD1.5 and SDXL models. It processes batch image inputs to generate images with similar facial features.
Install ComfyUI-Janus-Pro in Custom Node Manager Download Janus-Pro-1B or Janus-Pro-7B and place into models/Janus-Pro folder, so you'll have
Model card for CLAP: Contrastive Language-Audio Pretraining
This workflow uses kontext dev together with Deblur LoRA; the prompts in the workflow maximize consistency of a character’s appearance, but because the control can be too strong it often fails to high‐res the image....
Tip : Adjusting the denoise value in the BasicScheduler to somewhere between 0.5 and 1 can help you achieve the results you want.
This is a modified version of kijai's vace video outpainting workflow from his WanVideoWrapper. This workflow uses an image as input instead of a video, and that image will be animated and outpainted. You can also...
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
Its made for full body shots with one person. If there's multiple faces results will not be good. Also it will not work with portraits (face closeup) - will rather degrade quality.
