Скачать или смотреть New Workflow. Flux.2 Klein Head/Face Swap - UNLIMITED reference Edit

FEB 16 update: New finetuned f2k models (True v1) available:
Flux2-Klein-9B-True-Q8_0.gguf 9.74 GB https://huggingface.co/wikeeyang/Flux...
Flux2-Klein-9B-True-fp8.safetensors 9.08 GB https://huggingface.co/wikeeyang/Flux...
Flux2-Klein-9B-True-bf16.safetensors 18.2 GB https://huggingface.co/wikeeyang/Flux...

Sharing my NEW workflow. Thanks for your kind feedback. Please enjoy creating!

Workflow: https://drive.google.com/file/d/1lv7S...
Features:
Head & Face Swap Klein 4b & 9b
Unlimited reference images

All Model links in the workflow.
Nodes used in W/F:
ComfyUI-Easy-Use
ComfyUI-EditUtils
ComfyUI-KJNodes
Extra Models for ComfyUI
CRT-Nodes
ComfyUI-EditUtils

Shorts:    • ComfyUI - Loading a BIG list of prompts fr...  
Text Prompt Lists (for multi prompt image edit testing):
1. https://drive.google.com/file/d/1FqQg...
2. https://drive.google.com/file/d/1VoVP...

Buy me a coffee: https://buymeacoffee.com/jackie_there...

📝 Editing Notes

Software Used:
All images and videos are fully generated and edited using ComfyUI (locally on my PC).
Images and Videos are enhanced (if any) only in ComfyUI.
AI Models:
100% self-hosted models (no cloud services used). Models include text-to-image, image-to-video, and video enhancement pipelines.

Hardware Setup:
Locally generated on my own GPUs and CPUs, without outsourcing or third-party AI services.
CPU: Intel i5 12400f/14700K
RAM: 96GB/48GB
GPU-1: RTX 3090 OC 24GB
GPU-2: RTX 5060 Ti 16GB
GPU-3: RTX 3080 OC 12GB

Workflow Style:
Image creation (T2I / I2I) → Video generation (T2V) → Video enhancement/upscaling → Final editing.
No copyrighted or external stock content is included.

Editing Process:
Raw AI-generated clips combined, trimmed, and color-balanced.
Upscaling and enhancement applied for higher clarity.
Sound/music (if present) is original or properly licensed.

Ethics & Transparency:
All media shown is AI-generated by me using ComfyUI.
No scraping of copyrighted material.
Shared for educational, creative, and entertainment purposes.

Disclaimer:
These visuals are AI-generated concepts and not real footage. They represent experimentation with local generative workflows.