Google Veo 3
Veo 3 is an earlier generation of Google’s video model family, known for producing short clips with generated sound. The current official Veo pages describe Veo 3.1. Access depends on whether you use a consumer product, the Gemini API, or Google Cloud, so an old tutorial may no longer describe a working endpoint.
Current options
For browser-based video creation, see Google Flow and the Flow access guide. Flow exposes supported models through its own account and credit system; using the application is a different route from migrating an API integration.
What it offers
Google Cloud documents Veo 3.1 generation from text, images, and first and last frames, with audio, reference-image support, and video extension on supported models. These controls suit shot drafts, product scenes, and storyboard experiments. Available resolutions, clip lengths, and reference controls differ between the standard, Fast, and Lite variants.
Access and cost
The Gemini API shut down veo-3.1-generate-preview, veo-3.1-fast-generate-preview, and veo-3.1-lite-generate-preview on October 22, 2026. Its video guide directs developers to Gemini Omni Flash or GA models on Google Cloud. This closure does not mean that all Veo products have stopped. Google Cloud still documents the separate Veo 3.1 GA endpoints.
Practical limits
Consumer subscriptions and Flow allowances do not include unrestricted API use. Check the regional Google AI plan and the model offered inside the product before paying. Eligible subscribers may receive Cloud credits, but Cloud billing, covered services, and capacity conditions still require checking. Migration to Gemini Omni changes the model and integration flow; it is not a model-name substitution. Recheck supported inputs, duration, audio, safety rules, and costs, then review the generated clip before using it.