Back to News
Google DeepMind•June 30, 2026
Gemini Omni Flash
Freemium
AI
video generation
multimodal AI
video editing
### TL;DR
Gemini Omni Flash is Google DeepMind's latest multimodal AI model, engineered to fundamentally transform how users generate and interact with video. It excels in complex tasks requiring cohesive text-to-video synthesis, step-by-step conversational editing, and real-world physics reasoning, all supported by a massive 1M+ token context window. This model is particularly adept at seamless style transfers, multi-turn reference blending, and dynamic motion tracking, making it a highly capable engine for developers, creators, and advanced production workflows.
Key Insights & Metrics
Pricing
Input (text, image, video, audio): $1.50 / 1M tokens; Text output (response and reasoning): $9 / 1M tokens; & Video Output: $0.10 / second of video ($17.50 / 1M video output tokens)
Cost structure
Version
Flash
Current release version
Hardware
No specific hardware required
Compute requirements
Category
Freemium
Licensing model
Region
United States
Primary region
Key Features
- Multimodal input processing (text, images, audio, video)
- Conversational video editing
- Real-world knowledge integration
- SynthID watermarking for AI-generated content identification
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!