Video Generation with AI Gateway
By Steven Van ·
AI Gateway adds beta video generation for Pro and Enterprise plans, with four launch models across 17 variations.
AI Gateway can now generate video, not just text and images. The feature is in beta, available to Pro and Enterprise plans and paid AI Gateway users on Vercel.
- Videos can be generated programmatically with AI SDK 6's experimental_generateVideo, using the same API, authentication and observability dashboard as text and image generation.
- A no-code AI Gateway Playground is embedded on each video model's page, for comparing providers, tweaking prompts and downloading results.
- Four models are available at launch, in 17 variations: Grok Imagine from xAI (fast, strong instruction following, style transfer), Wan from Alibaba (reference-based generation, multi-shot storytelling, identity preservation across scenes), Kling (image-to-video and native audio, with 3.0 adding multishot video and automatic scene transitions), and Veo from Google (high visual fidelity, physics realism, native audio).
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →