Gemini Omni 1.1 Flash

Use the public model ID below for both Official and Ext routes. Provider model names are mapped by the server.

gemini-omni-1.1-flash

Official: automatic 3–10 second output; omit duration. Supports reference images, first/last frames and video editing. Ext: specify 4, 6, 8 or 10 seconds; default 6. Both support 360p, 720p, 1080p and 4k, landscape or portrait.

Ext supports one first-frame image or one/three reference images with generation_type:reference. Ext does not support two-frame interpolation or continuation by task ID. Supplying duration selects an Ext-capable route.

POST https://apimaster.ai/v1/videos/generations · POST https://apimaster.ai/v1/videos

curl "https://apimaster.ai/v1/videos/generations" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-omni-1.1-flash","prompt":"A paper boat on a sunlit lake, gentle camera movement","resolution":"720p","aspect_ratio":"16:9"}'

Ext:

{"model":"gemini-omni-1.1-flash","prompt":"A paper boat on a lake","duration":6,"resolution":"720p","aspect_ratio":"16:9"}

model · prompt · resolution · aspect_ratio · image_urls · video_urls

Official: first_frame_image · last_frame_image · image_with_roles · metadata.task · extend_from_task_id

Ext: duration · generation_type · nsfw_check

Official continuation: extend_from_task_id must reference your successful Official task on the same channel; it excludes video_urls. Public task IDs are mapped privately. A reference video must be at most 10 seconds. Ext reference-video requests omit duration; their pricing differs from ordinary generation.

{"id":"task_...","task_id":"task_...","status":"queued"}

Poll every 5–10 seconds until completed or failed. Download with the same API key. Successful video links appear in the usage log.

curl "https://apimaster.ai/v1/videos/task_..." -H "Authorization: Bearer YOUR_API_KEY"
curl -L "https://apimaster.ai/v1/videos/task_.../content" -H "Authorization: Bearer YOUR_API_KEY" -o video.mp4

queued → in_progress → completed / failed

Google’s verified 720p output baseline is $0.10136/second (5,792 tokens × $17.50/million). Input costs $1.50/million tokens; text output costs $9/million. Seconds alone do not reproduce the complete official bill. Other resolutions and Ext duration tiers have route-specific prices.

An estimate is reserved at submission and reconciled against the final provider bill using frozen channel and account price rules. Official input usage can increase the final amount. Failed generation refunds the generation reservation. Model cards and consumption logs show the applicable price.

Google Gemini API

400: invalid parameters · 401: authentication · 402: balance · 503: selected route lacks the requested capability