Google's Gemini Omni 1.2 Flash model is now integrated and accessible through the Vercel AI Gateway. This integration allows developers to leverage advanced text-to-video generation capabilities using a single API key, eliminating the need for a separate Google account. The model is designed to produce videos ranging from 3 to 15 seconds in length, complete with native audio. Additionally, it supports editing existing video clips, also between 3 and 15 seconds, by incorporating up to six reference images.
Developers can extend the duration of generated videos by adding 3 to 10 seconds per subsequent call, reaching a total maximum length of 30 seconds. The service also enables refinement of previously generated clips through follow-up calls, utilizing a previousInteractionId for continuity. Output options include various resolutions such as 360p, 720p, 1080p, and 4K, available in both 16:9 and 9:16 aspect ratios. The generated video content can be retrieved as a file by employing the generateText or streamText functions within the AI SDK, simplifying the process of incorporating dynamic video into applications.
Why it matters for businesses and developers
For businesses, this integration provides a simplified pathway to generate dynamic video content from text prompts, potentially enhancing marketing efforts, product demonstrations, or user engagement features. Developers gain streamlined access to a powerful AI model through Vercel's AI Gateway and AI SDK, reducing integration complexity and accelerating the deployment of advanced AI-driven video capabilities with a single API key.
This brief is our own summary. Read the original story, “Gemini Omni 1.2 Flash now available on AI Gateway”, at Vercel (published 5 Nov 2026).