Skip to main content

​
Overview

Google Veo 3.1 is the current GA video-generation model from Google, available on RouterBase. It generates 720p / 1080p video at 4–8 second durations with optional Gemini-synthesized audio. Two tasks are exposed: text-to-video and image-to-video.

​
Endpoint

This is an async endpoint. The POST returns immediately with {id, status: "pending"}; poll GET /v1/videos/generations/{id} until status is success (with data: [{url}]) or failed.

​
Playground

Try Google Veo 3.1 live without writing code:

​
Supported Tasks

  • image-to-video — animate a still image from a text prompt. The active model is also reachable via the google/veo-3-1-i2v-novita model id.
  • text-to-video — generate a clip from a text prompt only. Reachable via google/veo-3-preview-t2v-novita; the id retains the legacy preview slug, but the underlying upstream is the GA Veo 3.1 text-to-video endpoint.

​
Pricing

Both tasks price along the same two dimensions — duration_seconds and generate_audio — so the per-call cost depends on which tier the request lands in. (Text-to-Video on this model is fixed at 8s; the text-to-video form hides the duration control accordingly.)

​
Authentication

All requests require a Bearer token — create one at routerbase.com/api-key.