API preview
Qwen Image 2.1 Image Generator & Editor
Prepare text-to-image, image editing, and ordered multi-reference workflows in one Qwen Image 2.1 workspace.
Qwen Image 2.1
Official provider identity
API preview
Try the prepared text and ordered-reference workflow locally. No file is uploaded and no generation request is sent until a hosted API contract and pricing are verified.
0/5,000
Prepared pipeline defaults
- Output
- 1024 px
- Steps
- 40
- Guidance
- Off
- References
- Up to 3 in this preview
These are documented local Diffusers defaults. Hosted endpoint parameters and pricing remain pending verification.
Qwen Image 2.1 cannot upload files, create tasks, or charge credits while its production switch is off.
Workspace preview · not a model output
Generation and image-conditioned editing with multiple visual references. This illustration is not presented as a live output from this model.
Recipe: Preserve the designer sneaker silhouette, borrow the iridescent woven material palette, and use the urban pavilion geometry for one editorial footwear campaign.
mode: preview · references: 3 · ratio: 16:10 · api: pending
Generate and edit
One documented pipeline covers prompt generation and image-conditioned editing.
Ordered references
Add and reorder several local reference images to prepare a multi-reference request.
Visual conditioning
The published Diffusers integration describes Qwen3-VL visual conditioning for reference-aware creation.
What it does
The complete generation and editing workspace is prepared, but no capability, endpoint, or price is presented as confirmed before the provider publishes its contract.
How it will launch
- Confirm the official model name and provider endpoint.
- Map and contract-test the published parameters.
- Set pricing, run a bounded paid test, then enable it.