# Wan 3

Wan 3 in Sloosh: text, start-and-end-frame or reference video with sound, 2 to 30 seconds at up to 1080p, in six aspect ratios.

Last updated: 2026-10-15

**Key facts**

- Wan 3 is Alibaba's video model, and in Sloosh it makes 2 to 30 second clips from text, a start frame, or references.
- Wan 3 offers 480p, 720p and 1080p in Sloosh.
- Wan 3 takes up to 10 reference images in Sloosh, or a start frame with an optional end frame.
- Wan 3 makes sound by default, and you can turn it off.
- Wan 3 has a seed setting in Sloosh to repeat a result.

Web page: https://sloosh.ai/docs/wan-3

Wan 3 makes video with native sound from text, from a start frame and an optional end frame, or from reference images. It runs up to 30 seconds in one take.

## When to use it

- Long single takes, up to 30 seconds, without stitching clips.
- Scenes guided by several reference images at once, such as a product and a setting.
- Animating a start frame towards an end frame.

## How to prompt it

- Name references by their position: "the subject in Image 1 walks past the building in Image 2". Images are numbered from 1 in the order you add them.
- Describe the action, the camera and the sound in the order they happen.
- For a frame-to-frame clip, describe the motion between the two frames, not what each frame already shows.

## Settings in Sloosh

- Aspect ratio: Auto, 16:9, 4:3, 1:1, 3:4 or 9:16.
- Resolution: 480p, 720p or 1080p.
- Length: 2 to 30 seconds.
- References: up to 10 images.
- Start and end frames: start frame optional; an end frame needs a start frame.
- A start or end frame can't be combined with reference images.
- Sound: on by default; you can turn it off.
- Seed: available, to repeat a result.
- Prompt: up to 20,000 characters.

Open it in [Video](https://sloosh.ai/studio/video) and pick Wan 3 in the model chip. The Generate button shows the cost first.
