Use Qwen Image 3.0 when bilingual prompt writing, explicit exclusions, repeatable seeds, and a structured set of output controls matter to the image workflow.
The final cost is rounded up to whole credits.
Example output
Qwen Image 3.0
Text to Image · Image to Image
Example output
Qwen Image 3.0
Best for
Supports
Try this model
No results to display yet.
Model examples
Review examples to help decide whether the model fits your task.
Model strengths
Model strengths
Suggested uses
Tasks to consider
Model details
Provider
Alibaba
Category
Image
Capabilities
Text to Image · Image to Image
Credit charges
The credit cost depends on your inputs and settings. Check the estimate before you start.
Inputs
Prompt + optional images
Prompt requirements
Qwen Image 3.0 FAQ
Qwen Image 3.0 fits Chinese-first or English image briefs, bilingual poster concepts, product compositions, and controlled visual experiments where negative prompts, prompt expansion, or a repeatable seed are useful.
Yes. Add up to three JPEG, PNG, WebP, BMP, GIF, or TIFF references, with a 10 MB limit for each image. With no reference it uses text-to-image; with references it uses image-to-image.
When enabled, prompt expansion rewrites a simple positive prompt before generation. Turn it off when the wording is already precise and you want the submitted instruction to stay as direct as possible.
Use the negative prompt to name visual elements or failure modes you want to avoid. Use an integer seed from 0 to 2,147,483,647 when you want a relatively stable basis for comparing controlled parameter changes.
A prompt-only output starts at 5 credits. Reference images can raise the final credit amount, so keep the input set focused on references that add distinct information to the brief.
Prepare a Chinese or English prompt, choose 1K or 2K, set the frame and output format, decide whether to expand the prompt, and optionally add negative prompt, seed, or up to three references.
Yes. Leave the reference area empty for text-to-image, or upload one to three references when the task needs subject, style, or composition guidance.
Choose 1K or 2K and the aspect ratio first. Then decide whether prompt expansion supports the brief, and add a negative prompt or seed only when you have a clear reason to control those parts of the result.
The model details are public. Uploading references and submitting a generation require sign-in so Rivya can validate files, use credits, and keep task history.
Stay here for a focused first test. Open Studio for saved variants, repeated seed comparisons, asset management, or a longer design workflow.
Compare alternatives
GPT Image 2.5 Flare supports image creation from a prompt and edits with image references, plus aspect ratio, resolution, and background settings.
Why consider it
Credit charges: Initial generation reserve: 1K 3 credits · 2K 10 credits · 4K 16 credits.
GPT Image 2.5 Sunburst generates images from a prompt, with optional reference images and separate aspect ratio, resolution, and background settings.
Why consider it
Credit charges: Rivya reserves 3 credits for 1K, 10 for 2K, or 16 for 4K before submitting the request.
GPT Image 2 on Rivya supports text-to-image generation and reference-image edits, with aspect ratio and resolution controls.
Why consider it
Credit charges: Initial reserve: 1K 3 credits · 2K 5 credits · 4K 8 credits.
Up to 800 characters
Developer access
Qwen Image 3.0 can be called through Public API v1. Modes that use reference media require uploads through the Files API. Check the model reference for supported inputs.