Google has released Gemini Omni 1.1 Flash, a new artificial intelligence model designed to generate professional-quality video content from text, images, audio, and existing footage. According to official company announcements, the tool expands on the initial Gemini Omni capabilities introduced at Google I/O 2026, aiming to replace traditional editing software with conversational workflows.
Scene Expansion and Context Handling
The primary feature of Gemini Omni 1.1 Flash is its scene-expansion capability, which allows the AI to take an existing video and extend it up to 40 seconds cohesively. Unlike earlier iterations that analyzed only the final frame, Gemini Omni 1.1 Flash analyzes the final 10 seconds of source material to maintain narrative consistency, motion dynamics, and physical realism. Google notes that the model generates these extensions in increments of up to 10 seconds rather than all at once.
Conversational Editing and Prototyping
The system lets users modify videos through natural language prompts while keeping characters and scenes recognizable. To reduce project costs and turnaround times, creators can generate low-resolution 360p prototypes to test ideas before committing to final renders. Beyond text-to-video and text-editing, the model also accepts a first and last frame as reference points to generate the intervening clip automatically.
Pricing and Resolution Scaling
Gemini Omni 1.1 Flash supports upscaling existing clips to higher resolutions, including 1080p and 4K. According to Google’s release details, pricing scales by resolution and duration:
- 360p: 3 centavos de dólar por segundo creado
- 720p: 10 centavos de dólar por segundo
- 1080p: 15 centavos de dólar por segundo
- 4K: 30 centavos de dólar por segundo
Availability and Verification
The tool is currently rolling out for subscribers of Google AI Plus, Pro, and Ultra tiers through Google Flow, the Gemini app, YouTube Shorts, and YouTube Create, alongside API access via Google AI Studio. To address trust and safety concerns in media generation, Google includes SynthID watermarking technology to identify AI-generated video across Google search, Chrome, and the Gemini ecosystem.
Worth a look