Pika H3 Flex
Generate short videos with synchronized audio using text, endpoint images, and mixed media references.
Gemini Omni 1.2 Flash
deepmind.google
Generally available video model supports generation up to 15 seconds, conversational editing, improved reference consistency, lower video output pricing, and higher paid-tier limits.
Visit website
Pricing
Freemium
Available on
Web, Gemini API, Google Cloud
Category
AI Video Tools
What is Gemini Omni 1.2 Flash?
Gemini Omni 1.2 Flash is Google DeepMind's model for generating and conversationally editing video, producing high-resolution clips with audio. It serves creators, production teams and developers building video applications.
Access routes include Google AI Studio, the Gemini API, Google Cloud and Google Vids. The Gemini app rollout targets Google AI Ultra subscribers.
Key features
Short video generation
Generate and edit clips up to 15 seconds long.
Conversational edits
Change backgrounds, products, lighting, weather or aspect ratio through follow-up instructions.
Reference guidance
Guide generation with images, motion clips up to three seconds long and audio cues.
Keyframes and extension
Guide sequences with first and last keyframes or extend existing clips.
Who is it for?
Marketing teams developing ad variants
Explore different backgrounds and formats for a short product video.
Creators planning scenes
Use reference images and keyframes to explore the look and movement of a short sequence.
Developers building video applications
Prototype a generation workflow before integrating it into a larger product.
How to use Gemini Omni 1.2 Flash
- Choose an available Google application, Google AI Studio or Google Cloud.
- Describe a clip and add references supported by that interface.
- Generate the clip, then request conversational edits.
Gemini Omni 1.2 Flash pricing
Freemium
Google Vids offers generation at no additional cost with a Google account or Google Workspace subscription. Google Cloud uses metered token billing; other applications have separate subscription and eligibility terms.
Google Cloud usage
USD 1.50 per million input tokens; USD 9.00 per million text-output tokens; USD 12.00 per million video-output tokens
Metered usage, charged per million tokens consumed.
Applies to Gemini Omni 1.2 Flash on Google Cloud. Video-output token consumption varies by resolution, and upscaling uses separate token accounting.
What to know before you use it
- Changing people's speech during video editing is currently restricted.
- Fast sports action, rapid script changes and complex physics remain areas for improvement.
- Gemini app video generation requires a Google AI subscription and age 18 or above, with features varying by tier and geography.
Frequently asked questions
Does it generate audio with video?
Yes. The model produces video with audio.
Is it available through an API?
Yes. Distribution channels include the Gemini API and Google Cloud.
Does every Cloud clip cost the same?
No. Charges depend on token consumption, including resolution and any upscaling.