gemini-omni-1.2-flash
A video generation and editing model from Google, supporting clips up to 15 seconds with synchronized audio, multimodal references, conversational edits, and extension for product videos, ads, and social content.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
Must be one of: gemini-omni-1.2-flash.
gemini-omni-1.2-flash 3–10 seconds for interpolation and extend. Omit for edit, upscale, or automatic mode with videos. Defaults to 5.
3 <= x <= 15Upscale requires an explicit 720P, 1080P or 4K target. Defaults to 720P.
1080P, 360P, 4K, 720P Omit for edit, extend, upscale, or automatic mode with videos. Defaults to 16x9.
16x9, 9x16 Leave unset for automatic mode.
edit, extend, image_to_video, reference_to_video, text_to_video, upscale First frame URL or Base64 data URI.
This field's allowed values depend on: first_frame_image.
Edit and extend accept at most two references. Upscale accepts none.
10Edit, extend, and upscale require one source video.
3Required except in upscale mode.
1Response
Business response. code=0 means the task was accepted; validation failures can return HTTP 200 with code=400.