How do I control how strong the illustration style looks?

Use the LoRA strength in LoraLoaderModelOnly as your primary control. If the node exposes guidance or edit-strength parameters on the Qwen Edit node (eba40a3a-…), adjust those as secondary controls. Increase LoRA strength for bolder style; reduce it to keep more photographic detail.

Can I make only stylistic changes and keep content intact?

Yes. Write instructions that focus on style, color, and rendering (e.g., “flat shading, soft pastel palette”) and avoid adding or removing objects. This nudges the model to preserve composition and subject while changing the look.

Does this workflow support region-specific edits or masking?

This graph is designed for global, instruction-based edits. For targeted regions, use a masking-enabled variant of the Qwen Edit workflow or add a mask path and a compatible edit node that accepts masks.

Qwen Image Edit for Illustration v 1.0

Q: Which models do I need and where do I get them?

Download the Qwen Image/Qwen Image Edit weights and the illustration LoRA from the Comfy-Org/Qwen-Image_ComfyUI and Comfy-Org/Qwen-Image-Edit_ComfyUI Hugging Face repos (see links in the workflow notes). Point the model loader node to the base weights and LoraLoaderModelOnly to the LoRA file.

Qwen Image Edit for Illustration v1.0 is a text-guided image editing workflow that turns your existing images into clean, stylized illustrations while preserving structure. It combines the Qwen Image Edit model with an illustration-focused LoRA so you can apply style changes (e.g., "convert this product photo into a flat vector illustration") without redrawing from scratch. The core path is simple: LoadImage brings in your source, PrimitiveStringMultiline captures your edit instruction, the model loader node (c7f5c302-e46e-4a0f-91bd-0aec474e7659) loads the Qwen Image/Qwen Image Edit weights, LoraLoaderModelOnly merges the illustration LoRA into the base model, the edit node (eba40a3a-f6c5-48ac-b58e-55525d06b373) runs inference, and SaveImage writes the result.

Technically, the LoRA is injected on-the-fly by LoraLoaderModelOnly, biasing the Qwen Edit model toward illustrated outputs while keeping the underlying content aligned with your source image. The edit node consumes the input image plus your instruction and uses the merged model to generate a new frame that follows your prompt. The included MarkdownNote provides in-graph tips, and all required model files are available from the Comfy-Org/Qwen-Image_ComfyUI and Comfy-Org/Qwen-Image-Edit_ComfyUI Hugging Face repos linked in the author notes.

API

Use this workflow from code

Every Comfy workflow is a JSON graph. The payload below is this workflow, exactly as ComfyUI runs it — fetch it from the URL, keep it in version control, or load it in ComfyUI and run it node by node.

847f17f3f95c.json

Fetching workflow JSON…

Run it from TypeScript or Python with the Comfy SDK. The same code targets Comfy Cloud or a ComfyUI you host yourself — only the base URL changes.

// Install (beta)
npm i @comfyorg/sdk

// Run this workflow (TypeScript)
import { Comfy } from "@comfyorg/sdk";

const client = new Comfy({ apiKey: "comfyui-..." });
const wf = await client.workflows.fromFile("workflow_api.json");
const job = await client.run(wf);
await job.getOutputs("<output-node-id>")[0].toFile("output.png");

The SDK takes a workflow in API format: open this workflow in ComfyUI and use File → Export Workflow (API).

Comfy Cloud API access requires a plan with an API key.

SDK docs Get an API key

FAQ

Frequently Asked Questions