Model index
Google DeepMind·Veo / Gemini Video

Gemini Omni Flash

Google’s Gemini-native model for cost-efficient video generation and conversational editing.

CurrentVideo generationMultimodalWeights not released
MODEL NUMBER162
Release date
2026-06-30
Organization
Google DeepMind
Family
Veo / Gemini Video
01

Specifications

Only published specifications are shown.

Input modalities
Text · Image · Video
Output modalities
Video · Audio
Open weights
No
Status
Current
02

Overview

Overview

Google’s Gemini-native model for cost-efficient video generation and conversational editing.

03

Changes from predecessor

Compared with Veo 3.1

PreviousVeo 3.1 · ctx
CurrentGemini Omni Flash · ctx
New capabilities Conversational editing Video generation Efficiency
04

Architecture & capabilities

Key capabilities

Conversational editingVideo generationEfficiency
05

Release history

API launch

Gemini Omni Flash opened video-generation API access.

View source
06

Related models

Select a card to view the related model.

07

Sources