This a very simple workflow automates the process of swapping a single source face onto a directory of target images. This is only meant for a single person face swap, not for any image with multiple people. This focuses only on face not hair or other parts of body. It uses the ReActor engine for high-quality face replacement and KJNodes for efficient batch folder processing. This does not require any memory heavy Safetensor or GGUF checkpoints, Clip files (Encoders), VAE or subgraphs or messy "spider web" like collection of numerous nodes and connections to do simple and fast batch face swap. This will work with almost 95 to 96 percent of images. This way you can perform batch face swap for most of the images for a LORA training Dataset for any unique consistent character you may wanna train but only for more complicated face swaps (for example - images with very complex emotions, partially visible face etc.) you can use workflows that use QWEN, KONTEXT etc. Using this is dead simple, you type the path to the folder where you saved your images you wished to do faceswap, then load your desired face reference image for your own unique consistent character. Then press run, that's it.. you will get your batch faceswapped outputs saved inside a folder named Batch_FaceSwap (this folder is located inside your ComfyUI's output folder where ComfyUI usually saves your outputs). One thing you have to make sure that your input images are not too big otherwise on low end GPUs (8 GB, 12 GB etc) the workflow might fail. Each of your ideal images is recommended to have 1024 pixels ( 1024x1024 pixels 1:1 Square, 1024x1536 pixels 2:3 Portrait &amp; 1536x1024 pixels 3:2 Wide Photo are the best ones ). I included a simple python script with this workflow that you can run on terminal to quickly resize your input images before faceswap but if you want a visual tool to do the same you can use my LORA Data Tool (you can find it here - <a target="_blank" rel="ugc" href="https://civarchive.com/models/2219046">https://civarchive.com/models/2219046</a> ). That LORA Data tool can also do automatic captioning and manual image caption auditing. Before you run this workflow ensure that you have the following installed via ComfyUI Manager for this workflow to function properly:* ComfyUI-ReActor (by Gourieff)* ComfyUI-KJNodes (by kijai)* ComfyUI-Image-Saver (by alexopus)I also provided tips on how to enable batch NSFW faceswap on a text file as well as on notes inside the workflow. Hopefully this will help you

Files with SHA256: 8f78423e43af6c3e8b532dfe6f26fc00b54cdd7438be5244879cb0b333cb38fb