Related: Google’s Gemma 4 passes 200 million downloads in just over two months.
Google has released two new AI models built around the same idea, make images and video fast and cheap enough to churn out at scale. In a June 30 announcement, Google introduced Nano Banana 2 Lite, the fastest and most cost efficient image model in its Nano Banana family, and pushed Gemini Omni Flash, its multimodal video model, to a wider release. The pairing is aimed squarely at developers and businesses that want to generate a lot of visual content quickly rather than craft a single perfect frame.
Nano Banana 2 Lite is the headline of the two. Google says it can produce an image in about four seconds, a low latency that makes it suited to workshopping ideas and generating many variations in quick succession rather than laboring over one shot. It is positioned as the cheap, high volume workhorse of the lineup, sitting below the general purpose Nano Banana 2 and the more powerful Nano Banana Pro, and it now replaces the original Nano Banana, which Google has relabeled a legacy model. On price it is meant to undercut, landing at about 3.4 cents per image, which Google frames as its most cost efficient image generator and which independent benchmarks have placed near flagship quality for a small fraction of the usual cost.
The second model handles motion. Gemini Omni Flash, first shown at Google’s developer conference in the spring, is a natively multimodal system that turns text prompts and images into short video and lets a user refine the result through plain language, conversational editing across multiple turns, swapping characters, changing camera angles, or relighting a scene. Google put video output at ten cents per second, cheap enough to make iterating on clips practical rather than precious. Alongside it the company showed a demo app called Omni Product Studio, which takes still images generated by Omni and turns them into what Google describes as cinematic e commerce videos, a pointed hint at where it thinks the money is.
The strategy behind releasing the two together is about the full pipeline rather than any single model. As Google put it, developers can now connect rapid image generation to video creation and editing in one flow, drafting a picture with Nano Banana 2 Lite and animating or extending it with Omni Flash. Both models landed at once through Google AI Studio, the Gemini API, and Google’s Gemini Enterprise Agent Platform, with consumer surfaces following. Meeting developers inside the tools they already use, and pricing the models to be run in bulk, is how Google turns a flashy demo into something teams actually build products on.
The commercial logic is not subtle. Google has leaned into marketing its image and video tools as practical aids for making advertisements and marketing content, the kind of high volume, low stakes output where speed and cost matter more than artistry. That framing sits inside a louder argument about AI generated media, where a wave of what critics call AI slop has drawn real backlash even as companies keep investing heavily in the tools that produce it. Google is also tightening its ties to Hollywood, having recently struck a reported seventy five million dollar deal with the indie studio A24, a partnership that drew sharp criticism from parts of the creative community.
That tension is the backdrop for why these two models matter. Making generation faster and cheaper does not resolve the debate over whether the world needs more machine made images and video, it accelerates it, lowering the barrier for anyone to produce a flood of content. For developers and businesses, the appeal is straightforward, a cheap image model and a cheap video model that plug into the same pipeline and the tools they already use. For everyone else, the release is another marker of how quickly synthetic media is becoming a commodity, priced by the second and the thousand, and how much of the coming argument will be about what all that cheap content is actually for.
Related on Entrelligence: Google’s Gemma 4 passing 200 million downloads, AI cutting movie production costs, and Cursor’s iOS app for AI coding agents.
Your First 10 AI Skills
10 practical AI skills, copy-paste prompts and a 7-day plan to start using AI with confidence.
Download the guide →