Skip to content
WorkflowListedExternal
Open source

SDXL Portrait-to-Scene Master Workflow — v1.1

Added perspective control for subjects and environments, plus consistent pose tracking for characters. Removed the end-of-pipe SD Upscale.

Sourcecivitai.com/models/2491184
TaskMultiCapabilitiesText to image · UpscalingFamilyQwen-ImageVersionRelease 2818886 — Draw Separately 1.1.jsonEstimated VRAMNot yet characterizedBgrade

Runtime profile

Access
External workflow
Status
Listed
Version
Draw Separately 1.1.json
Source
Civitai
Updated
Sep 30, 2026

Source description

Added perspective control for subjects and environments, plus consistent pose tracking for characters. Removed the end-of-pipe SD Upscale.

Base styling remains driven by SDXL.

What you need

Model files
21.9 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

15

Node packs

4

  • comfyui_essentials

    Found

    Node pack · Registry

  • comfyui_layerstyle

    Found

    Node pack · Registry

  • comfyui-rmbg

    Found

    Node pack · Registry

  • was-ns

    Not resolved

    Node pack · Registry

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • Use SDXL Portrait-to-Scene Master Workflow to create scenes in a consistent SDXL artistic style from independently generated character and background elements, with manual character positioning and scaling and a blended-scene output.

Flow

  • Depth and LineArt extracted from the character guide background synthesis.

  • Qwen-VL blends images, checks for anatomical and logical issues, and supports targeted inpainting repairs.

Inputs and outputs

  • Provide independently generated character and background elements; the workflow produces a blended scene.

Setup

  • This is listed as a Workflow using Qwen as its base model, published by Enomoto Takane.

Dependencies

  • Named node packs: comfyui_layerstyle, comfyui_essentials, comfyui-rmbg, and was-ns.

  • Named model files: Qwen2.5-VL-7B-Instruct-Q3_K_S.gguf; illustriousXLDepth_v20.safetensors; z_image_turbo_example.safetensors; Qwen-Rapid-AIO-NSFW-v19_Q6_K.gguf; mmproj-Qwen3VL-4B-Instruct-F16.gguf; qwen-4b-zimage-heretic-q8.gguf; Qwen3VL-4B-Instruct-Q4_K_M.gguf; and waiIllustri­

  • Additional named model files: noobaiXLControlnet_openposeModel.safetensors; z_image_turbo-Q6_K.gguf; ultramix_balanced.safetensors; qwen-image-edit-2511-multiple-angles-lora.safetensors; qwen_image_vae.safetensors; Qwen-Image-Edit-F2P.safetensors; and illustriousXLLineart_v10.saf

Configuration

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

    Suggestion · not verified

Cautions

  • Before using it, verify that each named model file and node pack is available under the intended filename.

    Suggestion · not verified

  • No license is specified in the listing; check the applicable usage terms before using it.

    Suggestion

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiSourceVersion 2818886
civitai.com/models/2491184

Verified Sep 30

Source excerpts

3 excerpts
NoteSep 30civitai.com/models/2491184

Source context: 228 downloads · Type Workflows · Base model Qwen

NoteSep 28civitai.com/models/2491184

Source context: 227 downloads · Type Workflows · Base model Qwen

NoteJul 18civitai.com/models/2491184

Source context: 149 downloads · Type Workflows · Base model Qwen