Concept image
“A bright cafe interior with teal chairs, natural window light, and a simple poster on the wall.”
Check room geometry, lighting, and poster legibility.
Qwen Image 2.1 AI Image Generator and Editor
Create an image from a prompt, or revise an existing image with up to three image inputs in Magic Hour. Qwen Image 2.1 is a fast choice for everyday concepts and edits.
Start in the image generator for a new concept. To edit an existing image, open the image editor and select Qwen Image 2.1.
Run with API:
Fast, easy, and scalable
How to evaluate Qwen Image 2.1
A practical decision guide based on Magic Hour's current product support and the questions creators ask when comparing models.
Best fit on Magic Hour
Quick image concepts and prompt-led edits using a source image with optional additional references.
Consider another model when
Choose a video model when a sequence or camera movement is required.
Three useful first-run tests
Use the prompts below to test the failure modes that matter before committing to a larger batch.
Source edit
“Keep the subject's pose and clothing; change the background to a quiet city park at golden hour.”
Compare the subject and edges with the original image.
Two references
“Use the first reference for the product shape and the second for the color palette; make a clean catalog image.”
Check whether shape and colors follow the intended references.
These prompts are evaluation recipes, not controlled benchmark results. Run the same prompt and settings across models before drawing a conclusion.
Trusted by teams at
How to create with Qwen Image 2.1
Move from a prompt to an image, then refine it in the editor if needed.
- 1
Describe your image
Write the subject, composition, style, and details you want to create. - 2
Choose a resolution
Select an available resolution and output count for your plan. - 3
Generate or edit
Review the result. For a targeted change, open it in the image editor and describe the edit.
Where Qwen Image 2.1 fits
Use it for quick text-to-image concepts and prompt-led revisions. Editing takes a limited number of image inputs and returns one result per run.
Fast concepts
Generate variations from a text prompt at the resolution your plan supports.
Prompt-led edits
Use a source image and additional inputs to guide a focused revision.
Qwen Image 2.1 — Model Card
Key specs, capabilities, and limitations.
Cost
Generation and editing — 640px / 1K / 2K = 10 / 10 / 20 credits per image
Plans
free, creator, pro, business
Editing inputs
Source image plus up to 2 additional images, including saved references
Frequently asked questions
Qwen Image 2.1 is a fast option for creating images from text and making prompt-led edits to an existing image. It suits quick concepts and revisions when you want both workflows in one model.
Supported resolutions: 640px (from free), 1K (from creator), 2K (from creator). The available options and total credit cost appear before you generate.
The editor accepts up to 3 input images total, including a source image and any saved references. Editing returns one image per run.
One image costs 10 credits at 1K or 20 at 2K. The Create form shows the total for the selected resolution and output count before generation.
