Wan 3.0

Wan 3.0

Wan 3.0 unifies text, first-frame, first-and-last-frame, and mixed-reference video generation in one model. On MachGen it creates 2-30 second clips at 480p, 720p, or 1080p, supports Adaptive plus five fixed aspect ratios, generates audio in the same request, and accepts up to 10 images, 5 videos, and 5 audio references.

Wan 3.0 video generation for long-form directed motion, adaptive framing, synchronized sound, and ordered image, video, and audio references.

Why choose Wan 3.0

Key strengths

• One model spans text-to-video, image-to-video, first-and-last-frame transitions, and mixed-reference generation.
• Output can run from 2 to 30 seconds at 480p, 720p, or 1080p, with a Smart duration option when the exact length should follow the request.
• Reference mode accepts up to 10 images, 5 video clips, and 5 audio clips while preserving their cross-media order.
• Synchronized audio can be generated with the picture, while Adaptive framing lets the model derive composition from the supplied material.

Good fit for

• Cinematic one-take scenes with sustained motion and a clear beginning, middle, and end.
• Product and character shots anchored by visual identity, movement, voice, ambience, or music references.
• Controlled transitions where both the opening and closing composition are supplied.
• Longer social, advertising, and pre-visualization sequences that need native sound and several coordinated reference types.

Inputs and output settings

This table reflects the model routes and controls currently exposed by MachGen.

Media, File, or Webpage?

These inputs solve different problems. Media gives Wan 3.0 visual or audio direction. File and Webpage give it material to read and turn into a video. Choose one mode per request.

Use File or Webpage in the Playground

1. Choose Reference to Video and select Wan 3.0.
2. Choose File or Webpage above the reference area.
3. Upload one document or paste one public HTTPS webpage URL.
4. Describe what Wan should extract, what video it should create, and which facts it must not invent. Add the desired structure, camera style, sound, and length when they matter.
5. Select the output settings, review the estimate, and generate.

File prompt example: Read the creative brief as the factual source. Create a concise launch video that introduces the problem, product solution, and final brand shot. Preserve the product name and stated claims; do not invent prices or specifications.

Webpage prompt example: Use the webpage as the factual source for a clean product explainer. Introduce its main capabilities, preserve names and stated facts, and omit claims that are not present on the page.

You do not need an @name for File or Webpage. The prompt is optional for these modes, but a clear brief gives the model a much stronger structure.

API request shape

{
  "model": "Alibaba-Wan-3.0",
  "task_type": "R2V",
  "prompt": "Turn the brief into a factual 10-second product launch video.",
  "src_file_urls": ["https://example.com/creative-brief.pdf"],
  "video_config": {
    "duration_secs": 10,
    "height": 1080,
    "aspect_ratio": "adaptive",
    "audio": true
  }
}

For a webpage, replace src_file_urls with src_webpage_urls. Direct file and webpage URLs must use HTTPS and be publicly accessible. Do not send media arrays, src_file_urls, and src_webpage_urls together.

Using Wan 3.0

1. Choose a supported generation mode in Playground.
2. Add the required prompt and source or reference assets.
3. Select output settings from the controls shown for that mode.
4. Generate, then inspect the returned asset before reusing the settings through the API.

Practical guidance

• Describe the subject, action or composition, environment, and visual direction in a clear order.
• For image animation, describe the intended motion and camera behavior instead of repeating what is already visible.
• Give each reference a clear role in the prompt, using the names inserted by the Playground.
• Use compatible start and end frames, then describe the transition between them.
• When audio is enabled, include any dialogue, ambience, or sound cues that matter to the scene.

How to get better Wan 3.0 results

• Write the shot as a timeline: establish the subject and environment, then action, camera movement, interaction, and sound cues in order.
• Name each reference in MachGen and use its @name in the prompt; state whether it controls identity, motion, lighting, voice, ambience, or rhythm.
• Use Adaptive when the source framing should lead. Pick a fixed ratio when delivery composition matters more than matching the references.
• For first-and-last-frame work, describe the physical transition and continuity constraints instead of restating the two images.
• Stress-test hands, object contact, reflections, occlusion, and audio timing before extending a result into a longer sequence.

Limits and availability

• MachGen admits only the modes and settings listed above for this catalog model.
• Source assets must finish uploading before a request can be submitted.
• Generated results can vary between requests, including when the same prompt and settings are reused.
• Wan 3.0 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

Learn more

For model background, technical details, and architecture, see the official Wan 3.0 creation guide.

Try this model in the MachGen Playground.

Official documentation may describe capabilities or parameters that are not currently exposed by MachGen. Use the Inputs and output settings section above as the source of truth for this page.

Wan 3.0 pricing

See Pricing for currently published MachGen rates. When available, the Playground estimate reflects the selected task and output settings.

Wan 3.0 on MachGen

Deployment: Pass-through. Wan 3.0 is pass-through. MachGen sends the request through its upstream provider integration and returns the generated asset.

The Playground and API sections on this page list the inputs and output settings exposed for this model.