Veo on Vertex AI model versions
Stay organized with collections Save and categorize content based on your preferences.

Veo on Vertex AI is a multimodal model that supports creating videos from text or image (Preview) prompts.

The veo model and version are the following:

Model name	Version
Veo 2 (GA allowlist)	`veo-2.0-generate-001`

The following table describes limits and modalities available:

Limits	Value
Modalities	text-to-video generation image-to-video generationPreview
API calls (prompts per project per minute)	10
Request latency	Videos are typically generated within a few minutes, but may take longer during peak usage.
Maximum number of videos returned per request	4
Maximum video length	8 seconds
Supported returned video resolution (pixels)	720p
Frame rate	24 frames per second (FPS)
Aspect ratio	16:9 - landscape 9:16 - portrait
Maximum image size uploaded or sent in a request (image-to-video generation) Preview	20 MB

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2025-04-01 UTC.