SDXL Portrait-to-Scene Master Workflow

Use SDXL Portrait-to-Scene Master Workflow to create consistent SDXL-style scenes from independently generated characters and backgrounds. You can position and scale the character, then blend the elements into an integrated scene.

The workflow automatically extracts Depth and LineArt from the character to guide background synthesis.

Qwen-VL supports image blending, automated self-checking of anatomical and logical issues, and targeted inpainting repairs.

Inputs are independently generated character and background elements; the output is a blended scene.

The workflow lists Qwen as its base model and Enomoto Takane as its publisher.

Listed node packs are comfyui_layerstyle, comfyui_essentials, comfyui-rmbg, and was-ns.

Listed model files include illustriousXLDepth_v20.safetensors, illustriousXLLineart_v10.safetensors, mmproj-Qwen3VL-4B-Instruct-F16.gguf, noobaiXLControlnet_openposeModel.safetensors, qwen_image_vae.safetensors, qwen-4b-zimage-heretic-q8.gguf, Qwen3VL-4B-Instruct-Q4_K_M.gguf, and

Additional model files are qwen-image-edit-2511-multiple-angles-lora.safetensors, Qwen-Image-Edit-F2P.safetensors, Qwen-Rapid-AIO-NSFW-v19_Q6_K.gguf, Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf, ultramix_balanced.safetensors, waiIllustriousSDXL_v160.safetensors, z_image_turbo_example.s.s?

Use the manual canvas to adjust the character’s position and scale before background synthesis.

Suggestion · not verified

Before use, verify that every listed model file and node pack is available with the intended filename.

Suggestion · not verified

No license is specified, so verify the applicable usage terms before publishing work made with the workflow.

Suggestion