# Qwen-Image Edit

Canonical page: https://fuser.studio/models/qwen-image-edit

Qwen-Image Edit in Fuser changes an image from a text instruction and optional references, keeping what you don't mention. Add LoRAs for style.

## Specs

| Spec | Value |
| --- | --- |
| Inputs | Prompt (text), Images (up to 4 images), Negative Prompt (text) |
| Output | image |
| Cost | 52 credits per run |

Instruction-driven multi-image editing for e-commerce staging, apparel mockups, and cinematic storyboards without manual masking

Creator: Alibaba Group

Also known as: Qwen-Image-Edit-Plus

[Edit with natural language](https://app.fuser.studio)

## How Qwen-Image Edit works

Execute advanced multi-image staging, element removal, and portrait adjustments in three intuitive steps without painting manual selection masks.

### Upload Source Assets

Upload your source image, headshot crop, or pair of assets for apparel mapping and product integration.

### Set Operation & Prompt

Pick your target operation—such as adding a background, transferring a shirt graphic, or expanding a portrait—and describe the desired scene.

### Generate Cohesive Edit

Let the specialized vision-language LoRA generate a high-fidelity edit with cohesive lighting, shadows, and contours.

## What Qwen-Image Edit is good at

Specialized LoRA modules handle complex multi-image compositing, surface warping, and scene synthesis through natural language instructions.

### Automated Apparel Mapping

Transfer 2D graphic files onto photographed garments using the dedicated shirt-design LoRA, wrapping artwork across dynamic wrinkles, fabric drape, and ambient creases with realistic lighting response.

### Contextual Product Staging

Isolate foreground products from plain studio backdrops and merge them into fully rendered environments complete with matching directional light, surface ambient occlusion, and floor reflections.

### Multi-Subject Portrait Merging

Composite two to four separate individual headshots into a single, cohesive group portrait that balances unified color grading, matching camera focal length, and natural proximity.

### Cinematic Storyboard Progression

Advance existing frames into sequential narrative progressions using the Next Scene trigger, directing camera pans, tilts, and atmospheric shifts while preserving character and environment continuity.

## Made with Qwen-Image Edit

Explore studio-grade product integrations, editorial fashion mockups, and atmospheric cinematic continuations generated across specialized editing modes.

### E-commerce cosmetic staging with natural surface reflections

### High-fashion editorial apparel mockup with dynamic fabric draping

### Atmospheric product commercial hero frame

### Cinematic next-scene storyboard progression

## What people build with Qwen-Image Edit

From e-commerce merchandisers and apparel brands to cinematographers and portrait photographers, see how creators streamline high-touch compositing workflows.

### E-Commerce Product Staging

Replace sterile white studio backdrops with textured, contextual lifestyle scenes that calculate authentic contact shadows and surface reflections around consumer products.

### Apparel & Merchandise Mockups

Map flat 2D artwork and typography onto apparel photography, automatically conforming designs across natural cloth folds, wrinkles, and ambient highlights.

### Cinematic Storyboard Continuity

Advance single concept frames into consecutive narrative storyboards by prompting directional camera moves and environmental shifts while maintaining visual continuity.

### Editorial Portrait Expansion

Expand cropped headshots or tight portraits into fully composed professional editorial images with custom wardrobes, postures, and atmospheric backdrops.

### Advertising Asset Production

Clean up distracting background elements, balance harsh studio lighting, and assemble multi-person composite shots without tedious manual masking.

## Frequently Asked Questions

### Is Qwen-Image Edit the same as Qwen-Image-Edit-Plus?

Yes, Qwen-Image Edit is built directly on the Qwen-Image-Edit-Plus and Qwen2.5-VL 20-billion parameter vision-language architecture developed by Alibaba Group. It operationalizes this foundation with specialized LoRA modules optimized for specific image tasks—including background staging, garment graphics mapping, headshot expansion, and cinematic frame continuity.

### What is the difference between Add Background and Integrate Product?

Choose **Add Background** when you have an isolated subject on a plain or white background and want the model to generate a brand-new environment from a text prompt. Choose **Integrate Product** when you already have two specific images—a product and a distinct background photograph—and need the model to blend the product into that target scene with matching 3D perspective, reflections, and contact shadows.

### Which operation should I use for apparel mockups versus headshot expansion?

Use the **Shirt Design** operation when mapping a 2D graphic onto clothing, providing the flat artwork as the first image and the model photo as the second image while using the prompt trigger `put this design on their shirt`. For transforming a tight face crop into an upper-body or full-body editorial image with custom styling, select the **Face to Full Portrait** operation.

### How do I generate cinematic frame continuations with Next Scene?

Select the **Next Scene** operation and start your prompt with the mandatory prefix `Next Scene:` followed by your intended camera motion or environmental evolution. For the smoothest cinematic transition, set the `effect_strength` slider between 0.7 and 0.8 to balance narrative progression with visual scene continuity.

### How many images can I upload at once?

The model accepts up to 4 images depending on the selected task. **Shirt Design** requires exactly 2 images (the flat graphic and the apparel photo), **Group Photo** accepts between 2 and 4 individual headshots to blend into a unified composition, and single-subject operations like **Add Background** or **Remove Element** take 1 image.

### What are the model's main limitations?

The model is not designed for pixel-perfect virtual try-on where intricate garment stitching or exact proprietary patterns must match 1:1, nor does it generate images purely from text without input imagery. It also struggles with micro-adjustments to individual fingers and ultra-fine latent inpainting, where manual brush masks remain more effective.

## Try Qwen-Image Edit on Fuser

Instruction-driven multi-image editing for e-commerce staging, apparel mockups, and cinematic storyboards without manual masking

[Edit with natural language](https://app.fuser.studio)

## Related models

- [Gemini Image](https://fuser.studio/models/gemini-image.md)
- [GPT Image](https://fuser.studio/models/gpt-image.md)
- [Seedream](https://fuser.studio/models/seedream.md)
- [FLUX.3 Image](https://fuser.studio/models/flux-3-image.md)
- [FLUX.2](https://fuser.studio/models/flux-2.md)
- [Topaz Image Upscale](https://fuser.studio/models/topaz-image-upscale.md)

Node reference: [Qwen-Image Edit docs](https://docs.fuser.studio/docs/nodes/image/qwen-image-edit.md)

## More articles

- [Best AI Virtual Try-On Models (2026)](https://fuser.studio/articles/best-ai-virtual-try-on.md)
- [AI Background Removers Tested: Bria RMBG 2.0 vs Rembg (2026)](https://fuser.studio/articles/best-ai-background-removers.md)
- [Qwen Image Edit Guide: Text Edits, Object Edits and Camera Angles](https://fuser.studio/articles/qwen-image-edit-guide.md)