You have an outfit. A pose. A background. You want all three merged into one cohesive image without shooting it, without compositing it manually, without three separate edit passes.
Reference your images by number. Describe the scene. Get one composed result out.
Run it now on Floyo!
How It Works
Upload your reference images. In your prompt, refer to each one by number "image 1," "image 2," "image 3." Describe what you want the final scene to look like. Qwen Image Edit 2509 reads all references simultaneously and composes them into one output.
No masking. No layering. No compositing software. One prompt, one result.
The 2509 update specifically improves multi-image alignment elements sit together more realistically, lighting matches better across inputs, and complex combinations resolve more cleanly than the original Qwen version.
Key Inputs
Your Reference Images
Upload the images you want combined. Each gets a number you reference in the prompt.
Works well with:
fashion garments and outfit combinations
product shots you want placed together in one scene
pose or body reference combined with an outfit or character
background or environment images to place subjects into
any visual elements you want merged into one cohesive composition
Prompt
Reference each image explicitly by number and describe the final scene.
Structure that works:
"Place the person from image 1 wearing the outfit from image 2, standing in the location from image 3, natural lighting, editorial style"
More examples:
"Combine the jacket from image 1 with the trousers from image 2 on the model from image 3, studio lighting, fashion lookbook""Place the product from image 1 and the product from image 2 together on the surface from image 3, clean product photography, white background""Use the pose from image 2 with the character design from image 1, background from image 3, cinematic lighting""Merge the handbag from image 1 with the outfit in image 2, editorial fashion photo, soft daylight"
Be specific about what comes from which image. The more clearly you reference each input, the more accurately the model places them.
What This Is Great For
Fashion and outfit combinations: Merge garments, accessories, and poses from separate reference images into one preview shot. Test outfit combinations before committing to a photoshoot.
Product composite showcases: Place multiple products together in one scene for catalog pages, campaign imagery, and collection displays.
Character and costume design: Combine character design references with pose references and environment references into a single composed illustration.
Mockup generation: Drop a product into a lifestyle scene or a garment onto a model without a shoot. Fast enough to generate multiple combinations in one session.
Creative concept work: Merge visual references from different sources into one cohesive scene for moodboards, pitch decks, and concept presentations.
What to Watch Out For
Always reference images by number in your prompt. Describing what you want without explicitly calling out "image 1," "image 2," and so on gives the model no clear mapping between your inputs and your intent. Explicit numbering is what makes multi-image compositing accurate.
Large style differences between reference images very different lighting, color temperature, or image quality make coherent compositing harder. References that were shot in similar conditions or have compatible aesthetics produce cleaner results.
Very complex scenes with four or more inputs can lose coherence. Start with two or three references and add a fourth only when the simpler combination is working well.
If the output doesn't match your intent, the prompt is almost always the fix. Add more specific placement instructions, more explicit references to each image number, and clearer descriptions of lighting, framing, and how the elements relate to each other.