Google's Gemini Omni 1.2 Flash lands on AI Gateway: one API key now generates video with native audio
Synopsis
Google's Gemini Omni 1.2 Flash is now available through AI Gateway, so developers can call it with one API key and no separate Google account. The model generates 3–15 second videos with native audio from text prompts and edits 3–15 second source clips using up to six reference images; each call extends a video by 3–10 seconds up to a 30-second total. Output comes in 360p, 720p, 1080p, and 4K at 16:9 or 9:16.
Interpretation
The video generation model Gemini Omni 1.2 Flash can now be called through AI Gateway, letting developers generate video from text prompts with a single API key instead of a separate Google account. Using Google's video generation previously meant going through Google's own account and access path; now the same gateway key reaches this model. The integration notes show calls made through generateText or streamText in the AI SDK, with the model name google/gemini-omni-1.2-flash, and the generated video returned as a file.
One call produces a 3–15 second video with native audio, and the model can also edit 3–15 second source clips with up to six reference images. Extension adds 3–10 seconds per call up to a 30-second total, and follow-up calls can refine a generated clip using previousInteractionId. Output covers 360p, 720p, 1080p, and 4K at 16:9 or 9:16; the example code generates a three-second 360p clip and saves it as video.mp4.
Perspective
These integration notes address developers calling models through AI Gateway and describe the availability, input and output specifications, and call pattern of Gemini Omni 1.2 Flash; they give no generation-quality results or comparison with other video models.
The durations, resolutions, reference-image count, and 30-second total listed in the notes are the specifications the model currently supports; the material does not say whether these will change with the version.