Video Generation with AI Gateway

AI Gateway now supports video generation through AI SDK 6, enabling programmatic video creation across four model providers with a unified API.

• •Four video model families are available: xAI Grok (fast, instruction-following with style transfer), Alibaba Wan (reference-based generation and multi-shot storytelling), Kling (image-to-video with native audio and 3.0 multishot support), and Google Veo (high visual fidelity, physics realism, cinematic lighting)
• •Five generation types are supported: text-to-video, image-to-video, first-and-last-frame interpolation, reference-to-video (consistent character identity), and video editing via style transfer
• •AI SDK 6 exposes a unified `experimental_generateVideo` API where model-specific capabilities are configured via `providerOptions`, supporting aspect ratio, duration, resolution, and audio settings
• •A no-code AI Gateway Playground is embedded in each model page for prompt experimentation and provider comparison without writing code
• •Curren