Google launches Gemini Omni Flash video model at $0.10/sec and Nano Banana 2 Lite image model into GA
Google Cloud has added two new generative-media models to its Gemini Enterprise Agent Platform. Nano Banana 2 Lite (formally Gemini 3.1 Flash-Lite Image) is now generally available, delivering image generation in as little as four seconds — the fastest and most cost-efficient model in the Nano Banana family. Gemini Omni Flash, a conversational video generation and editing model, enters public preview priced at $0.10 per second of video output.
Gemini Omni Flash is built around four headline capabilities: natural-language conversational editing (swap characters, relight scenes, change angles), multimodal inputs combining text, images, and video, physics-aware world simulation grounded in Gemini's knowledge base, and on-screen text/kinetic-typography rendering synced to action. Every video output includes natively generated audio. Features such as audio references, scene extension, and higher resolutions are flagged as "coming soon" via the API.
Nano Banana 2 Lite improves on its predecessor Nano Banana (Gemini 2.5 Flash Image) with better world knowledge for contextual scenes, cross-generation character consistency for storyboarding and e-commerce try-ons, and faster in-image text/localization rendering. Adobe is integrating both models into Adobe Firefly, WPP has already piloted Gemini Omni Flash inside its WPP Open agentic marketing platform, and Invideo is exploring the model's VFX and hybrid live-action use cases.