# Wan 3 API

> Make Wan 3.0 videos online: text, image or reference to video with sound, up to 30 s at 1080P. Wan 3.0 Prime and Wan 2.7 too. From 3 credits a second.

Website: https://wan3api.org · Full details for AI assistants: https://wan3api.org/llms-full.txt

Last updated: 2026-11-01

## What is Wan 3.0?

Wan 3.0 is the newest model in Alibaba's Wan video line (Tongyi Wanxiang, from Tongyi Lab). Its public beta opened on August 6, 2026, and Alibaba launched it officially on August 24, 2026. It generates 2 to 30 seconds of video in one pass at 480P, 720P or 1080P, with voices, effects and music made at the same time.

Wan 3.0 is not open source, so you use it through Alibaba Cloud or a service like this one. This site runs Wan 3.0, its premium tier Wan 3.0 Prime and the previous Wan 2.7 in the cloud: write a prompt, add a first or last frame, or give it reference images, clips and audio, then download the MP4. Developers get the same models over a REST API.

### Wan 3.0 at a glance

- Made by: Alibaba's Tongyi Lab, as part of the Wan (Tongyi Wanxiang) model line. Model id: wan3.0-video. (source: [Alibaba Cloud Model Studio](https://modelstudio.console.alibabacloud.com/model-releases/wan3.0-video))
- Released: Officially launched on August 24, 2026, after a public beta that opened on August 6. (source: [Reuters](https://www.reuters.com/business/retail-consumer/alibaba-launches-wan30-ai-video-model-after-10-billion-share-sale-2026-08-24/))
- Length: 2 to 30 seconds in a single pass. (source: [Alibaba Cloud Model Studio](https://modelstudio.console.alibabacloud.com/model-releases/wan3.0-video))
- Output: 480P, 720P or 1080P, in 16:9, 9:16, 1:1, 4:3 or 3:4. (source: [Alibaba Cloud Model Studio docs](https://www.alibabacloud.com/help/en/model-studio/text-to-video))
- Inputs: Text; a first frame, or first and last frame; or references: up to 10 images, 5 video clips and 5 audio clips (15 s of video and 15 s of audio in total). (source: [Alibaba Cloud Model Studio](https://modelstudio.console.alibabacloud.com/model-releases/wan3.0-video))
- Sound: Voices, effects and music generated in the same pass as the picture. (source: [Alibaba Cloud Model Studio](https://modelstudio.console.alibabacloud.com/model-releases/wan3.0-video))
- Weights: Not released. The last open-weight Wan model is Wan 2.2 (Apache 2.0, 2025). (source: [GitHub Wan-Video/Wan2.2](https://github.com/Wan-Video/Wan2.2))
- Official app: Also available in Alibaba's own Wan app. (source: [wan.video](https://wan.video/))

Wan 3 API is an independent service and is not affiliated with or endorsed by Alibaba. Wan and Wan 3.0 are trademarks of their owners.

## Create a video

Generator: https://wan3api.org/create

1. Describe the shot: Say what happens: subject, action, camera, light and sound. Or start from an example.
2. Add pictures or references: Optional: a first or last frame, or reference images, clips and audio to follow.
3. Generate and download: Pick the model, resolution and length, see the price, and download the MP4 when it's ready.

## When Wan 3 API is a good fit

- You want Wan 3.0, Wan 3.0 Prime or Wan 2.7 in the browser or over a REST API, without setting up an Alibaba Cloud account.
- You want to try it free: a new account gets 15 credits (one 5-second 480P Wan 3.0 video).
- You want to pay per second of video with one-time credit packs (300 credits for $9.99), no subscription, and see the exact price before each render.
- You need text to video, first/last frame or reference mode (up to 10 images, 5 clips and 5 audio clips) with sound generated in the same pass.
- Not a fit: you need unlimited free renders, clips longer than 30 seconds, document or web-page input, or open weights to run yourself.

## Models and tools

- [Wan 3.0](https://wan3api.org/wan-3-0): Alibaba's newest model: up to 30 s with sound.
- [Wan 3.0 Prime](https://wan3api.org/wan-3-0-prime): Alibaba's premium Wan 3.0 tier, for final renders.
- [Wan 2.7](https://wan3api.org/wan-2-7): Previous generation: 720P or 1080P, 2–15 s.
- [Text to video](https://wan3api.org/text-to-video): Describe a shot and get a video with sound.
- [Image to video](https://wan3api.org/image-to-video): Animate a picture, or move from one image to another.
- [Reference to video](https://wan3api.org/reference-to-video): Images, clips and audio guide character, style and voice.

## Models and prices

### Wan 3.0

- Length: 2–30 s, with sound
- Modes: text to video, first/last frame (image to video), reference images, video and audio
- 480P: 3 credits per second of video (≈ $0.10 at the smallest pack)
- 720P: 6 credits per second of video (≈ $0.20 at the smallest pack)
- 1080P: 11 credits per second of video (≈ $0.37 at the smallest pack)
- Reference video: 3 (480P), 6 (720P), 11 (1080P) credits per uploaded second, on top of the output
- Reference images: included

### Wan 3.0 Prime

- Length: 2–30 s, with sound
- Modes: text to video, first/last frame (image to video), reference images, video and audio
- 480P: 5 credits per second of video (≈ $0.17 at the smallest pack)
- 720P: 9 credits per second of video (≈ $0.30 at the smallest pack)
- 1080P: 17 credits per second of video (≈ $0.57 at the smallest pack)
- Reference video: 5 (480P), 9 (720P), 17 (1080P) credits per uploaded second, on top of the output
- Reference images: included

### Wan 2.7

- Length: 2–15 s, with sound
- Modes: text to video, first/last frame (image to video), reference images, video and audio
- 720P: 6 credits per second of video (≈ $0.20 at the smallest pack)
- 1080P: 9 credits per second of video (≈ $0.30 at the smallest pack)
- Reference video: 6 (720P), 9 (1080P) credits per uploaded second, on top of the output
- Reference images: included

## API

REST API base URL: https://api.wan3api.org/v1. Authenticate with `Authorization: Bearer sk-…` (create keys at https://wan3api.org/api-keys). Renders are async tasks: POST /videos/generations, then poll GET /tasks/{id} or pass callback_url.

- [Wan Video API Quickstart](https://wan3api.org/docs): Create an API key, submit a Wan 3.0 text-to-video task, poll it and download the MP4 in three requests. cURL, JavaScript and Python examples.
- [API Authentication and Rate Limits](https://wan3api.org/docs/authentication): Authenticate with a Bearer API key, keep keys safe and stay within the per-key rate limit.
- [Create Video API: Text, Image and Reference to Video](https://wan3api.org/docs/video-generation): POST /v1/videos/generations: every request field, the text, frames and reference modes, idempotency keys and what a render costs.
- [Task Status API: Poll, List and Delete Renders](https://wan3api.org/docs/tasks): GET /v1/tasks/:id to poll a render, list recent tasks, delete finished ones, and the task object every endpoint returns.
- [Callbacks: Get Notified When a Video Is Ready](https://wan3api.org/docs/callbacks): Pass callback_url and we POST the finished task to your server: payload, retries and how to verify it.
- [Wan Video Models and API Pricing per Second](https://wan3api.org/docs/models): Wan 3.0, Wan 3.0 Prime and Wan 2.7 in the API: resolutions, lengths, modes and the price per second in credits and US dollars.
- [API Error Codes and Retries](https://wan3api.org/docs/errors): Every HTTP status and error code the API returns, what each means, and which ones are safe to retry.

## Pricing

- [Credit packs](https://wan3api.org/pricing): one-time packs from $9.99. Failed renders return their credits automatically.

## Guides

- [5 Photo Tips for Better AI Template Videos](https://wan3api.org/blog/photo-tips): How to pick photos that turn into convincing videos: lighting, framing, one face per photo and more.

## Privacy and policies

- Uploaded photos and generated videos are private to the account that made them.
- Only use photos of yourself or people who agreed to appear; photos of public figures are rejected.
- [Privacy policy](https://wan3api.org/privacy) · [Terms of service](https://wan3api.org/terms) · [Refund policy](https://wan3api.org/refund-policy) · [Cookie policy](https://wan3api.org/cookie-policy)
- [About](https://wan3api.org/about) · [Contact](https://wan3api.org/contact): support@wan3api.org