Google launched Gemini Omni 1.1 Flash on August 27, a video-generation model that outputs clips at up to 4K resolution, marking a step from preview-grade AI video toward production-ready output1,2,3.
The model upgrades Google's prior Omni 1.1, which topped out at 720p, by adding 1080p and 4K tiers alongside a faster, cheaper 360p preview mode. Google positions the model as a tool for developers and professional creators.
Three production-oriented features define the release. Scene expansion lets users continue generating frames from the end of an existing video in 10-second increments, up to a maximum clip length of 40 seconds. Camera-movement control accepts specified start and end frames to produce smooth transitions and professional-style camera moves. A video-reference input accepts clips of up to 3 seconds, allowing the model to maintain context and character consistency when building new scenes.
Google published a tiered pricing structure. The 360p preview tier costs $0.03 per second, runs up to 60% faster than the standard tier, and costs roughly one-third as much, making it suited for storyboard iteration. The 720p tier costs $0.10 per second, 1080p costs $0.15 per second, and 4K costs $0.30 per second.
ANALYSIS The layered pricing creates a deliberate pipeline: creators can iterate cheaply at 360p, then render final deliverables at 4K without switching tools or models. The per-second cost structure and the 40-second clip ceiling orient the model toward professional post-production workflows rather than casual consumer use.