Qwen Image 3.0
qwen3 / image-to-image

Qwen Image 3.0 is Alibaba's advanced image generation model for realistic images, structured layouts, rich visual details, and high-quality typography.

Model Type:
Pricing: Qwen image 3.0 Image-to-Image costs 4.8 credits per image (≈ $0.024) for 1K/2K output. Input images are charged at 0.5 credits per image. High-tier top-ups bring the effective price down to approximately 90% of the prices listed above.
Input

Click to upload or drag and drop

Supported formats: JPEG, PNG, WEBP, BMP, GIF, TIFF Maximum file size: 10MB; Maximum files: 3

Upload an image file to use as input for the API

The positive prompt that describes the image content, style, and composition you want to generate or edit. Both Chinese and English are supported.

0/5000

The output image resolution.

The output image size.

Generate image format

Whether to enable intelligent prompt rewriting. Default: true (recommended). When enabled, the model optimizes the positive prompt, which significantly improves results for simple descriptions.

The negative prompt that describes content you do not want to appear in the image.

0/5000

The random seed. Value range: [0, 2147483647]. A fixed seed helps maintain relatively stable results.

A configurable parameter. Defaults to true in the Playground.

There is no guarantee that everything can be filtered out; if you are not satisfied with the results, you will need to make your own arrangements.
Output
output typeimage
Generated output
Examples

Explore different use cases and parameter configurations

README

Complete guide to using Qwen Image 3.0

Qwen Image 3 API: Text-to-Image and Image Editing

Original image
ParameterValue
ModelsStandard: qwen3/text-to-image, qwen3/image-to-image. Pro: qwen3/pro-text-to-image, qwen3/pro-image-to-image
EndpointPOST /api/v1/jobs/createTask, then Get Task Details or a callBackUrl for the result
PromptRequired; Chinese or English; up to 5,000 characters per request. Per the Qwen team, the model handles long prompts of about 4.5K tokens
Negative promptOptional; up to 5,000 characters
Input images (image-to-image)1 to 3 image URLs; JPEG, PNG, WebP, BMP, GIF or TIFF; up to 10 MB each
Resolution1K or 2K (image-to-image defaults to 1K)
Aspect ratio1:1, 3:2, 2:3, 4:3, 3:4, 16:9, 9:16, 21:9 (default 16:9)
Output formatPNG or JPEG
Prompt rewritingprompt_extend, on by default
SeedOptional, 0 to 2147483647; random if omitted
PricingPer model and resolution, shown in the pricing card above
Long Prompt, Dense Layout: A Four-Panel Water Cycle Infographic
Small Text and Real Textures: A Handwritten Recipe Card on Wood
Multilingual Text: One Poster in English, Japanese and Spanish
Image Editing: Handwritten Notes Added to an Existing Photo
Qwen Image 3Qwen Image 3 Pro
Models on Kie.aiqwen3/text-to-image, qwen3/image-to-imageqwen3/pro-text-to-image, qwen3/pro-image-to-image
Request parametersSame as ProSame as standard
Artificial Analysis text-to-image score (Oct 2026)10751087
LMArena text-to-image score (Oct 7, 2026)Not listed1255
LMArena text-rendering score (Oct 7, 2026)Not listed1283
PricingShown in the pricing card aboveShown in the pricing card above

Qwen Image 3 is the Qwen team's model for text-to-image generation and image editing, offered on Kie.ai in standard and Pro versions. Its capabilities, parameters and examples are covered above.

Not on Kie.ai: each generation uses credits, as shown in the pricing card above. New Kie.ai users receive 80 free credits for testing.

As of October 2026, the Qwen team has not published Qwen-Image-3.0 weights on its Hugging Face page, so there is no official download for running it locally. The open-weight models there are earlier releases such as Qwen-Image 2.1. Kie.ai provides this model as an API with a playground; there is no app to download or install. Another open-weights image model is MiniMax H3 Image, which MiniMax describes as generating images of about 16 megapixels directly, without an upscaler.

Sign up on Kie.ai and create a key on the API Keys page (kie.ai/api-key), then send it in every request as the header Authorization: Bearer YOUR_API_KEY. This is a Kie.ai key, not an Alibaba Cloud key: one key and one credit balance work for every model on Kie.ai. Each of the four models has its own page in the Kie.ai docs: Qwen Image 3 text-to-image and image-to-image, Qwen Image 3 Pro text-to-image and image-to-image. Requests are asynchronous: createTask returns a taskId, and you get the result through your callBackUrl or by polling Get Task Details.

Qwen Image 3 costs 4.8 credits (about $0.024) per image at 1K or 2K. Qwen Image 3 Pro costs 6.4 credits (about $0.032) per image at 1K and 12 credits (about $0.06) at 2K. Text-to-image and image-to-image cost the same per output image; image-to-image also adds 0.5 credits (about $0.0025) per input image. Prices may be updated; the pricing card at the top of this page (select a model) always shows the current rate.

Both take the same request on Kie.ai; only the model name changes. The comparison table above shows their public leaderboard scores and when to start with each.

Qwen3 is the Qwen team's family of language models; image generation comes from the separate Qwen-Image series. On Kie.ai the Qwen Image 3 model names start with qwen3/, for example qwen3/text-to-image, and they run Qwen-Image-3.0.

Yes. The image-to-image models take one or more input images by URL plus an instruction, and return the edited image. The parameter table above lists the input limits, and the recipe card example shows an edit.