https://x.com/icekiubAI/status/2045116980206702691?s=20 A WAN 2.2 low noise character Lora is required for this to work but I think it's one of the better solutions out there right now and I quickly go over the whole process in the video This is mostly a faceswap pipeline, hair color, body and clothes will be from the original video, select your reference videos with that in mind. We use WAN 2.2 T2V not I2V make sure you get that right Workflow in the zip My character on the left in the samples The video for this will be coming today or tomorrow WAN 2.2 T2V: https://huggingface.co/icekiub/WAN-2.2-T2V-FP8-NON-SCALED/resolve/main/WAN2.2t2vLOWNOISEFP8.safetensors 4step: https://huggingface.co/lightx2v/Wan2.2-Lightning/resolve/main/Wan2.2-T2V-A14B-4steps-lora-250928/low_noise_model.safetensors text encoder: https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors vae: https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/blob/main/split_files/vae/wan_2.1_vae.safetensors

This heavily requested workflow will let you generate images with consistent rooms and backgrounds. Couple of disclaimers The workflow is based mostly on Klein and is not perfect Depending on the base image, it will sometimes generate gibberish. Either fix your prompt or change the base image. You will need a lora Klein lora if you want the best possible likeness to your character. If you do no want to train a lora, you can run other faceswapping solutions on the results you get I believe it's one of the better options out there, if not the only one - but let me know if you find better! The free version generate 3 angles or the room you chose as input (Including the generated one), while the sub version generates 6 or 7 in total, the rest is the same. You can easily expand the free version with different prompts. Models: Angles and final image gen: Klein : https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-fp8/blob/main/flux-2-klein-9b-fp8.safetensorsif if no access on HF: https://modelscope.cn/models/black-forest-labs/FLUX.2-klein-9b-fp8/file/view/master/flux-2-klein-9b-fp8.safetensors?status=2 text_encoders: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/text_encoders/qwen_3_8b_fp8mixed.safetensors vae: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/vae/flux2-vae.safetensors The room can be real or generated with whatever model you want. I used qwen so this is what I will provide: QWEN IMAGE 2512: https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_2512_fp8_e4m3fn.safetensors

This workflow will let create consistent photosets for social media content or really whatever type of content you want. This version made specifically for people but changing the prompt, it could be used for product photography for example! Klein KV: https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-kv-fp8/blob/main/flux-2-klein-9b-kv-fp8.safetensorsKlein No KV: https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-fp8/blob/main/flux-2-klein-9b-fp8.safetensorsif no access: https://modelscope.cn/models/black-forest-labs/FLUX.2-klein-9b-fp8/file/view/master/flux-2-klein-9b-fp8.safetensors?status=2 text_encoders: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/text_encoders/qwen_3_8b_fp8mixed.safetensors vae: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/vae/flux2-vae.safetensors

https://www.youtube.com/watch?v=XUijQ5Pc3xg This pipeline uses Z-image Base for initial generation and turbo for the final refinement. Also going over ZIB training in the vid. Links: All for ZIB: https://huggingface.co/Comfy-Org/z_image/tree/main/split_files All for ZIT: https://huggingface.co/Comfy-Org/z_image_turbo/tree/main/split_files *8 steps ZIB (ComfyUI one): https://huggingface.co/alibaba-pai/Z-Image-Fun-Lora-Distill/tree/main

Using this workflow, you should be able to get a good dataset to train your LoRA on, I have many video on how to do this locally and using runpod. All the prompts can be changed to generate the type of images you want, just make sure you have one prompt per line in the CR prompt list node Model: https://huggingface.co/black-forest-labs/FLUX.2-klein-9b-fp8/blob/main/flux-2-klein-9b-fp8.safetensors Text encoder: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/text_encoders/qwen_3_8b_fp8mixed.safetensors vae: https://huggingface.co/Comfy-Org/flux2-klein-9B/resolve/main/split_files/vae/flux2-vae.safetensors