# Wan 3.0

Thirty seconds in one take, and it takes pictures, clips and sound as references.

Model id: `wan-3.0`
Type: Video
Made by: Alibaba
Inputs: Text, First frame, References
Length: 2 to 30 s
Resolution: 480p, 720p, 1080p
Aspect ratio: adaptive, 16:9, 4:3, 1:1, 3:4, 9:16
Updated: Oct 12, 2026

Coming soon.

## Pricing

| Setting | MiniAPI | fal | Difference |
| --- | ---: | ---: | ---: |
| 480p | $0.038 / s | $0.05 | -24% |
| 720p | $0.075 / s | $0.10 | -25% |
| 1080p | $0.16 / s | $0.20 | -20% |

You are not charged for a job that fails.

## Parameters

| Name | Required | Values | Default | Description |
| --- | --- | --- | --- | --- |
| `prompt` | For text | text |  | What to make, in plain words. |
| `start_image_url` | For first frame | https URL |  | A picture for the first frame. Only with first frame. |
| `web_url` | No | https URL |  | Public webpage URL to base the video on. Requires enable_thinking=true. Only pages that do not require login can be read. Only with references. |
| `end_image_url` | No | https URL |  | A picture for the last frame. Only with first frame. |
| `file_url` | No | https URL |  | Document URL to base the video on. Requires enable_thinking=true. Only with references. |
| `reference_audio_urls` | No | https URLs |  | Sound to follow or lip-sync to. Only with references. |
| `reference_image_urls` | No | https URLs |  | Pictures that show a face, an outfit or a product to keep the same. Only with references. |
| `reference_video_urls` | No | https URLs |  | Clips that show a motion or style to follow. Only with references. |
| `resolution` | No | 480p, 720p, 1080p | 1080p | Output size. Higher is sharper, slower and costs more. |
| `aspect_ratio` | No | adaptive, 16:9, 4:3, 1:1, 3:4, 9:16 | adaptive | Shape of the result, for example 16:9 or 9:16. |
| `duration` | No | 2 to 30 | 5 | Length in seconds. The price changes with it. |
| `audio` | No | true, false | true | Include generated audio. |
| `enable_prompt_expansion` | No | true, false | true | Let the model add detail to a short prompt. |
| `enable_thinking` | No | true, false | false | Enable enhanced reasoning before generation. |
| `seed` | No | 0 to 2147483647 |  | The same seed with the same settings gives a similar result. Leave it out for a random one. |

Page: https://api.minicut.ai/models/wan-3.0
API reference: https://api.minicut.ai/docs/api