Veo 3.1 (Standard, Fast, Lite)
Google DeepMind's main Veo video model, with native audio and a broad set of filmmaking controls. It is available in the Gemini app, Flow, the Gemini API and Google's enterprise platform in Standard, Fast and Lite tiers. Veo 2 and…
Best for
Cinematic clips with native sound and fine creative controls such as reference images, first/last frame and scene extension.
Overview
Google DeepMind's main Veo video model, with native audio and a broad set of filmmaking controls. It is available in the Gemini app, Flow, the Gemini API and Google's enterprise platform in Standard, Fast and Lite tiers. Veo 2 and Veo 3.0 API models were shut down on 2026-06-30, and there is no official Veo 4.
Strengths and limits
- Native audio: dialogue, sound effects, ambience
- Ingredients-to-video, first/last frame, scene extension, object insert/remove
- 1080p and 4K output; three price tiers
- Now ranks below newer models (incl. Google's own Gemini Omni Flash) on Artificial Analysis
- Google notes natural short speech is still in development
Published benchmark scores
Scores are as published by the source; settings (reasoning effort, tools, attempts) differ between labs, so compare with care. How to read benchmarks →