Skip to main content
Veo
Veo generates video with native audio. This page uses the Gemini Developer API field definitions. MixRoute keeps model and prompt at the root and places both media inputs and generation settings directly inside metadata. POST https://api.mixroute.ai/v1/video/generations
Use a flat metadata object. Do not add metadata.instances or metadata.parameters. Veo 3 models generate audio natively; do not send generateAudio. Vertex-specific output storage, Pub/Sub, masks, and other enterprise-only settings are not part of this request format.

Models and Versions

Confirm route availability in the Model Marketplace.

Request Parameters

Scenario Constraints

  • Image objects use raw Base64 data, not a data:image/...;base64, prefix or an arbitrary URL string.
  • Reference images and video extension are Veo 3.1 capabilities; do not apply them to legacy Veo 3 routes.
  • Extension input must be a recent Veo output: 720p, 16:9 or 9:16, at most 141 seconds, and still within its retention window. Each extension adds 7 seconds, up to 20 extensions and 148 seconds total; durationSeconds must still be 8 for the extension request.
  • Frame rate is 24 fps and audio is generated natively. Output retention is limited; retrieve completed results promptly.

Examples

Set MIXROUTE_API_KEY before calling the API. Replace media placeholders with accessible inputs. Accepted requests create billable generation tasks; do not automatically resubmit after a submission timeout.

Python

Task Results

Save the MixRoute task ID returned by submission and poll Query Video Task. A successful submission creates a task; read the result only after the task reaches a successful terminal state. Response envelopes and result locations vary by route.