New
Qwen-Image 3AI image generator
Alibaba's newest image model, from 2026-07-21: images and edits of your own, with strong text in Chinese and English. On Leap, it sits beside 169 other models in one studio.
- Text to image
- Image edit
- 5 aspect ratios
- Start free, no card
- Every top model, one account
- Pay per run, priced before it starts
AI image generator
Qwen-Image 3 at a glance
Qwen-Image 3 is an image model from Alibaba. It makes images from words or up to 4 photos.
- Made by
- Alibaba
- Makes
- Images
- Starts from
- Words or up to 4 photos
- Shapes
- 1:1, 3:4, 9:16, 4:3, 16:9
- Per run
- Up to 4 images
- Prompt
- Up to 5,000 characters
- Price per run
- $0.0495
- On Leap since
- October 5, 2026
- Updated
- October 5, 2026
- 01Describe it, or start from a photo
- 02See the exact price, then press go
- 03Keep it in your library, or take it further
Pick the right one
Qwen-Image 3 and its siblings
| Model | Best for | Price per run |
|---|---|---|
| Qwen-Image 3 (this page) | Alibaba's newest image model, from 2026-07-21: images and edits of your own, with strong text in Chinese and English. | $0.0495 |
| Qwen-Image 2512 | Qwen-Image with sharper text, finer textures and more realistic people. | $0.022 |
Start with an idea
Three ways to try Qwen-Image 3
Side by side
More image models on Leap
No prompt needed
Or start from a recipe
Recipes write the prompt for you: add a photo, pick a look, and get a finished result.
Questions
Qwen-Image 3, answered
- What is Qwen-Image 3?
- Qwen-Image 3 is an image model from Alibaba. Alibaba's newest image model, from 2026-07-21: images and edits of your own, with strong text in Chinese and English. It is new on Leap and in preview: it works, and it may still change. On Leap it runs in one studio beside every other top model, with one account.
- Can Qwen-Image 3 start from a photo?
- Yes. Qwen-Image 3 starts from words or up to 4 photos.
- What resolution and sizes does it make?
- 1:1, 3:4, 9:16, 4:3 or 16:9.
- How is it different from Qwen-Image 2512?
- Qwen-Image 3: Alibaba's newest image model, from 2026-07-21: images and edits of your own, with strong text in Chinese and English. Qwen-Image 2512: Qwen-Image with sharper text, finer textures and more realistic people. They all run in the same studio, so you can try one prompt on each.
- How much does a run cost?
- An image costs $0.0495. You see the exact price before you press go, and a run that fails costs nothing.
- Is Qwen-Image 3 free to try?
- Yes. Sign up with a Gmail, Outlook or work address and you start with $2 of free credit, no card needed. The cheapest run is $0.0495 (an image), so the credit covers 40 of those. After that you pay per run, in dollars, with no subscription.
- Can I use what I make commercially?
- Yes. You own what you make on Leap, so you can use it commercially, and we never train AI on it. The people and brands shown in a result can still have rights in it; the Terms say more.
- Is there an API?
- Yes. Leap's REST API runs it as alibaba/qwen-image-3, at the same price per run as the studio, and agents can call it through Leap's MCP server. The request is at the bottom of this page.
- Do I need an Alibaba account?
- No. Leap sends your request to Alibaba's model and gives you what it makes. One Leap account runs Qwen-Image 3 and every other model, in the studio, through the API and from your agents.
Your idea, Qwen-Image 3.
Start free, no card. Every run shows its price before it starts.
Make it with Qwen-Image 3For developers
Run Qwen-Image 3 from code
- Model ID
- alibaba/qwen-image-3
- Makes
- Image
- Pricing
- Per run, quoted first: POST /v1/quotes
Inputs
What the model takes. Leave out anything with a default.
Promptprompt
Text, up to 5,000 characters, required
Imagesimages
Up to 4 imagesUploaded files to edit or combine, by ID.
Aspect ratioaspect_ratio
1:1, 3:4, 9:16, 4:3, 16:9. Default: 1:1
Imagesn
1 to 4. Default: 1
Enable Prompt Expansionsettings.enable_prompt_expansion
On or off. Default: onEnable automatic LLM prompt rewriting for better results.
Negative promptnegative_prompt
Text, up to 2,000 charactersWhat the image should not contain.
Seedseed
0 to 4294967295The same seed and input give the same image.
The request
The same envelope for every model; only the model and its input change. With prefer: wait=60 the answer holds the result when the run finishes within a minute; a longer one is read again by its ID. Agents can call it through the MCP server at app.tryleap.ai/api/mcp.
curl https://api.tryleap.ai/v1/generations \ -H "x-api-key: $LEAP_API_KEY" \ -H "content-type: application/json" \ -H "prefer: wait=60" \ -d '{ "model": "alibaba/qwen-image-3", "input": { "prompt": "A paper boat drifting down a rainy street at dusk" } }' # Video often outlasts the wait and answers "queued" or "running".# Put its "id" in GENERATION_ID and read it again until it is done:# curl https://api.tryleap.ai/v1/generations/$GENERATION_ID -H "x-api-key: $LEAP_API_KEY"