Veo 3.1 Fast Model Introduction

Model Overview

This is the high-speed generation edition of "Veo 3.1", the top-tier video generation model developed by Google DeepMind. It maximizes generation speed while maintaining high levels of video quality and consistency, making it ideal for prototyping and quickly visualizing ideas.

This model features an "Ingredients to Video" function that lets you upload multiple reference images to generate consistent videos. It outputs 8-second videos that retain the details of people and the textures of objects, while also reflecting cinematic camera work and physical simulations. Additionally, it supports native generation of audio such as environmental sounds and dialogue, as well as high-accuracy lip-syncing.

It also supports vertical videos (9:16 aspect ratio), enabling a powerful workflow for professional production scenarios that require both speed and quality, such as creating advertisement visuals, SNS content, and storyboards.

Pricing

Technical Specifications