omni-1.2-flash
AliasGemini Omni 1.2 FlashGemini Omni 1.2 Flash API by Google on TTAPI supports video generation from text or images, video editing, and continuation. Choose 360p to 4K output with per-second billing.
Gemini Video prompts, references, and results
What you can build with omni-1.2-flash
Text and first frame
Start from a text prompt or use one image to anchor the opening frame.
Reference media
Guide generation with up to ten reference images.
Edit or extend
Edit a source video or add 3–10 seconds to it using one public video URL.
Retrieve the finished video
Use the returned job ID to retrieve task status and the video URL, or supply a callback.
Per-action, in quota
Usage is metered in quota by model, action, and output settings.
Generation bills the requested duration. Continuation bills only the added seconds. Editing uses the source video's actual duration. Subtract 0.15 seconds, round up, and bill at least 1 second and at most 15 seconds. The request is rejected if the duration cannot be read or the source exceeds 15 seconds.
Headers and submit parameters
Use the endpoint below with your TTAPI key and the request fields shown.
/gemini/video/generationsSubmit a generation, edit, or continuation job and receive data.jobId.
Official reference/gemini/video/fetchPass the returned jobId as a required query parameter. Read the video URL at data.video_url when the job succeeds.
Official referenceHeaders
TT-API-KEYstringrequiredYour TTAPI API key.
Content-TypestringrequiredUse application/json.
Body
promptstringrequiredScene and motion prompt.
modelenumrequiredUse omni-1.2-flash.
aspect_ratioenumrequired16:9 or 9:16.
resolutionenumoptional360p, 720p (default), 1080p, or 4K.
durationintegeroptional3–15 seconds for text or image inputs, default 10. For extend, specify 3–10 added seconds, default 10. Omit duration for edit. The output follows the source video's length. Editing uses the source video's actual duration. Subtract 0.15 seconds, round up, and bill at least 1 second and at most 15 seconds. The request is rejected if the duration cannot be read or the source exceeds 15 seconds.
generation_typeenumconditionalframe requires one image. reference accepts one to ten images. edit and extend require exactly one video and no images. When omitted, one image is a first frame and two or more are references. A video requires edit or extend.
reference_imagesstring[]conditionalUp to ten HTTP(S) URLs. Base64 is unsupported. Image-based requests cannot include a video.
reference_videosstring[]conditionalExactly one publicly reachable HTTP(S) URL for edit or extend. The source duration must be readable for editing.
negative_promptstringoptionalDescribe content to exclude. An empty value is omitted.
hookUrlstringoptionalCallback URL for task completion or failure.
From key to first result
Copy the resolved endpoint, authentication headers, and request body for this model.
curl --request POST \
--url https://api.ttapi.org/gemini/video/generations \
--header 'TT-API-KEY: $TTAPI_KEY' \
--header 'Content-Type: application/json' \
--data '{"prompt":"A slow aerial shot above a misty lake at sunrise","model":"omni-1.2-flash","aspect_ratio":"16:9","resolution":"720p","duration":10}'