30s
Longer single generations
Create clips up to 30 seconds in one generation when the mode supports it.
Wan 3.0 · live on AIWan
Wan3.0 is Alibaba’s next-generation all-in-one video generation model built for multimodal creation. It supports text, image, video, audio, and structured references to produce more consistent, realistic, and controllable video content with enhanced storytelling and editing capabilities.
30s
Create clips up to 30 seconds in one generation when the mode supports it.
20
Combine up to 20 project references in a single creative task.
8
Use text, images, audio, video, documents, spreadsheets, slides, and web pages.
Wan 3.0
Instead of forcing every requirement into one long prompt, Wan 3.0 is built for a reference-led workflow. Describe the scene, add the visual, audio, motion, or project material that matters, and keep those references in context while you generate or edit.
For creators, that means fewer restarts when matching a character, product, environment, or art direction. You can move from concept frames to motion and revisions in a more connected workflow.
How it works
Create clips up to 30 seconds in a single generation with duration control
Why it helps
Spend less time stitching short clips and fixing jumps in identity, lighting, or pacing
How it works
Add up to 20 images, clips, audio files, or project documents
Why it helps
Keep the subject, brand direction, action, and context together instead of overloading one prompt
How it works
Use reference material to anchor characters, products, objects, and environments
Why it helps
Build ads, character content, and branded visuals with a more stable look across shots
How it works
Guide revisions with instructions and references
Why it helps
Change the scene, motion, styling, or environment without rebuilding the entire idea from zero
Start with the creation mode that matches your job, add only the references that matter, and iterate from the result instead of rebuilding every shot from scratch.
Wan 3.0 Studio
Wan 3.0 is most useful when a shot needs more than a good prompt: a recognizable product, a recurring character, a specific visual direction, or source material that needs to stay consistent through revisions.
Use product images, brand references, and motion direction to build campaign shots without repeatedly redefining the visual identity.
Anchor a character with reference images, then carry the same look into longer shots, new actions, and revised scenes.
Turn briefs, storyboards, source clips, and brand assets into variations for launches, social posts, trailers, and short-form campaigns.
Create with more control
Use image and video modes together so you can build stronger source material, animate it, and refine the result without leaving the creative flow.
Ask for targeted changes to motion, framing, styling, story beats, or environment instead of regenerating the whole concept.
Create or edit characters, products, compositions, and storyboards first, then use those visuals as a clearer starting point for motion.
Check text, numbers, logos, product details, and reference fidelity before using generated content in a final campaign or client deliverable.
Use more than a prompt
Add up to 20 references across eight content types so the model can work from the same brief, characters, product material, motion examples, and brand direction you are using.



AAC · 5.0s
30s
Longer single generations
20
Reference assets
8
Reference types
Creation modes
Open a supported video workspace and select Wan 3.0 directly from the model selector. You can switch modes without leaving the same Studio workflow.
Describe a scene, product, character, or visual direction and turn it into a source image.
Open workspaceUpload an image and guide changes to composition, styling, details, or visual direction.
Open workspaceDescribe the subject, action, camera, lighting, and mood to create a new video shot.
Open workspaceStart from a reference frame and add motion while keeping the core composition and subject recognizable.
Open workspaceCombine several visual references to guide identity, product details, environment, and overall art direction.
Open workspaceUse an existing clip as the starting point for guided edits, restyling, or scene changes.
Open workspaceUse these crawlable paths to move between the core generators and production prompt examples.
FAQ
Availability, pricing, API access, reference support, and generation limits in one place.
Yes. Wan 3.0 is officially live on AIWan. Open a supported video workspace, choose Wan 3.0 in the model selector, set your duration and resolution, add supported references when needed, and review the credit estimate before generating.
Wan 3.0 supports text-to-video, image-to-video, multi-reference generation, and guided video editing. It is especially useful when you need a product, character, environment, or visual direction to stay consistent across revisions; AIWan image tools can also help you prepare source frames before video generation.
The workspace shows the estimated Wan 3.0 credit cost before you submit a generation. The final cost can vary by mode, duration, resolution, audio, reference inputs, and other generation settings.
AIWan does not currently provide public Wan 3.0 API access from this Studio page. The browser-based Studio workflow is the current focus. If public API access is added later, authentication and usage documentation will be published separately.
Wan 3.0 supports a reference-led workflow across text, images, audio, video, documents, spreadsheets, presentations, and web pages, with up to 20 reference assets in a task. Use the most relevant references rather than adding material that does not affect the final shot.
Wan 3.0 supports clips up to 30 seconds in a single generation. The durations, resolutions, and options available for the selected mode are shown directly in the workspace.