DeepMind Launches Gemini Omni 1.2 Flash: 15s Video, Lower Cost, Better Consistency
This summary and analysis were generated by AI from the original article at DeepMind and may contain errors (how Viqus works). Read the source for full details.
7
What is the Viqus Verdict?
We evaluate each news story based on its real impact versus its media hype to offer a clear and objective perspective.
AI Analysis:
While the hype is moderate due to the incremental nature of model updates, the tangible improvements in consistency and workflow control represent a genuine, high-impact step for industry adoption.
Article Summary
DeepMind has launched Gemini Omni 1.2 Flash, making it generally available across the Gemini API and Gemini for business. This update significantly enhances video generation capabilities, notably extending clip length to 15 seconds and introducing multi-turn conversational editing. The model also boasts enhanced reference following, ensuring character, product, and style consistency across complex shots, such as virtual try-ons. Furthermore, the cost has been reduced by 30% from the previous version, positioning it as a powerful, cost-effective tool for commercial content creation across various Google platforms.
Key Points
- The new model supports native video generation and editing up to 15 seconds, a significant increase from previous versions.
- Gemini Omni 1.2 Flash offers multi-turn conversational editing, allowing users to refine clips iteratively without starting from scratch.
- The model improves reference adherence, maintaining consistency of characters, products, and styles across complex, continuous camera movements.
Why It Matters
This release represents a substantial maturation of generative video AI, moving it closer to production-grade, controllable workflows. The combination of longer duration, iterative editing, and cost reduction makes it a major tool for commercial advertisers and content creators. It signals a strong push by major labs to make high-fidelity video generation accessible and reliable for enterprise use cases, setting a new baseline for video AI tooling.