Gemini Omni 1.2 Flash adds 15-second video

Google has released Gemini Omni 1.2 Flash, a multimodal model that the company describes as faster and cheaper than its predecessor while adding native 15-second video generation and editing. The update is aimed at making video creation more accessible to developers and product teams, particularly those working with cost-sensitive workloads that need both speed and longer-form output. Native video generation is not new, but Google is emphasizing consistency across generated clips. That matters because short AI videos can quickly lose coherence between frames or scenes, which limits their usefulness in real products. The addition of native editing is also significant. Instead of only generating clips from scratch, users can refine existing output, suggesting a more iterative workflow for creative and commercial applications. Google says the model carries a 30% lower price, a detail likely to appeal to teams building video features at scale. Lower cost can make it practical to run more experiments, generate multiple variations or support higher-volume use cases. The combination of speed and reduced pricing positions Gemini Omni 1.2 Flash for workloads where latency and budget are key constraints. The release also reflects a broader trend: multimodal models are moving from text and images toward richer media generation and editing. For developers, the question will be how well the model handles real-world prompts, maintains visual consistency and integrates into existing pipelines. Google’s announcement provides a clear product direction, but adoption will depend on hands-on results. If the model delivers on its promises, it could lower the barrier for teams that want to add video creation and editing without building specialized infrastructure.