Wan 3.0 now available on AI Gateway
By Steven Van ·
Wan 3.0 from Alibaba is now available on AI Gateway as alibaba/wan-v3.0-video.
Alibaba's Wan 3.0 video model is now available on Vercel's AI Gateway as alibaba/wan-v3.0-video. It combines text-to-video, image-to-video, first- and last-frame conditioning, and reference-based generation, using image, video, or audio references, in a single model. This replaces the separate -t2v and -r2v model IDs that Wan 2.7 required.
The model generates clips up to 30 seconds at 30 fps, in 480p, 720p, or 1080p, with synchronized audio, up from Wan 2.7's limit of 15 seconds at 24 fps.
- Generation is asynchronous: a webhook can report when a render finishes instead of an HTTP request staying open for the whole render.
- References are passed through inputReferences, with a source and media type for each. Images accept hosted URLs or base64; video and audio references require hosted URLs.
- First- and last-frame conditioning takes one image per frame and can't be combined with other references.
Wan 3.0 can be tried in the model playground.
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →