Hey y'all!

Take an image of a character and extract the pose, expression, body type and style out as a character dummy!

This workflow is used in conjunction with my Lazy Character Fusion workflow, which also needs my Lazy Character Dummy workflow.

Special thanks to tori29umai (on Twitter) for the image2body lora and @DigitalPastel for workflows, models, and images I've used.

VERSION 1.0

• The workflow is based on the normal Qwen Edit 2509 workflow, same models too.

• In addition, it uses tori29umai's image2body lora model.

• The output format is a character dummy needed for my Lazy Character Fusion workflow, which also needs my Lazy Character Sheet workflow.

Input:

• Character Pose image

• Prompt (No need to adjust)

Output:

• Character Dummy image

• Character Dummy + Original info image

Models:

• Diffusion Model:

  • qwen_image_edit_2509_fp8_e4m3fn.safetensors

• LoRA;

  • Qwen-Image-Edit-2509-Lightning-4steps-V1.0-bf16.safetensors

  • QIE_image2body_V2_dim4_1e-3-000078.safetensors

• Text Encoder:

  • qwen_2.5_vl_7b_fp8_scaled.safetensors

• VAE:

  • qwen_image_vae.safetensors

This workflow is just a cleaned up and slightly modified version of the default ComfyUI Qwen Edit 2507 workflow. Its really quite simple. There's some additional info in the notes within the workflow.

Stay Cheesy Y'all!~

• Brie Wensleydale