WAN Video API: How to Use Alibaba's WAN 2.2 via API

What is WAN?

WAN is Alibaba's open-weight video generation model series designed for high-quality AI video creation. The WAN family includes WAN 2.1 and the newer WAN 2.2, both supporting text-to-video and image-to-video generation with strong motion quality and visual fidelity.

Key capabilities include text-to-video generation from natural language prompts, image-to-video conversion to bring still images to life, and animation/motion features through WAN 2.2 Animate. Because WAN is open-weight, the model weights are publicly available — but running inference at scale requires significant GPU infrastructure that Renderful handles for you.

How to Use the WAN API

With Renderful, accessing WAN models is as simple as making a single API call. Sign up for an API key, then send a POST request with your prompt and desired settings. Here's a working example:

curl -X POST https://api.renderful.ai/api/v1/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "alibaba/wan-2.2-14b",
    "input": {
      "prompt": "A golden retriever running through a sunlit meadow with wildflowers",
      "aspect_ratio": "16:9"
    },
    "webhook": "https://your-app.com/webhook"
  }'

The API is asynchronous — your webhook will receive the completed video once generation finishes. You can also poll the generation status endpoint for progress updates.

WAN Models Available

Renderful provides access to multiple WAN model variants through a single API. Choose the model that best fits your use case:

WAN API Pricing

Renderful offers straightforward pay-per-generation pricing for WAN API access with no monthly minimums or commitments.

Frequently Asked Questions

Related Articles