Service standards
What happens to a render from the moment you start it, what counts against your plan, and what each model delivers. The Terms are the binding contract; this page explains them in plain words.
Renders run in the background. The Studio shows each one with its place in the queue, an estimated start and its progress, and the API reports status and progress through GET /v1/tasks/{id}. You can close the browser at any time.
A render holds one unit of your plan allowance while it runs and is counted only when it finishes. If it fails or you cancel it, the unit goes back automatically. A failed Omni 1.1 Flash generation is never billed.
Your Veo 3.1 Lite and Nano Banana allowance resets with each paid month. Unused allowance does not roll over. Cancel any time; the plan keeps working until the end of the paid month.
Each API response carries a request ID (x-request-id) and each render has its own ID. Send either to support and we can find exactly what happened.
Times depend on the current queue. The Studio shows live wait and render times on the Queue page.
| Model | Output | Billing | Runs |
|---|---|---|---|
| Veo 3.1 Lite | MP4, about 8 s, 720p, 16:9 or 9:16 | 1 video from your plan | Queued, background |
| Omni 1.1 Flash | MP4, about 8 s, 720p, 16:9 or 9:16 | $0.15 per successful video (add-on) | Queued, background |
| Nano Banana 2 / 2.1 / Pro | Image, 16:9 9:16 1:1 4:3 3:4, 1 to 4 per request | 1 image from your plan per output | Queued, background |