Google packages Nano Banana 2 Lite and Gemini Omni Flash for teams counting regenerations

The most revealing part of Google’s new AI media launch is the spreadsheet it quietly hands to product teams. Google’s launch details put Nano Banana 2 Lite at about 4 seconds for a 1K image and $0.034 per image, which turns image generation from demo theater into a cost line a PM can model before lunch. Gemini Omni Flash, meanwhile, enters public preview for high-quality video generation and conversational editing, according to Google Cloud. That split matters because the expensive part of AI media products is rarely the first impressive output. It is the thirteenth regeneration, the A/B test nobody budgeted for, and the user who treats every prompt box like a slot machine.

The price tag is

the product Google Cloud says it is adding Nano Banana 2 Lite and Gemini Omni Flash to the Gemini Enterprise Agent Platform, according to a post by Michael Gerstenhaber, VP of Product Management for Cloud AI. Nano Banana 2 Lite, also called Gemini 3.1 Flash-Lite Image, is generally available, while Gemini Omni Flash is in public preview. Google Cloud describes Nano Banana 2 Lite as the fastest and most cost-efficient image generation and editing model in the Nano Banana family. That is not just a model claim. It is a packaging decision aimed at teams that need throughput more than trophy outputs. The use cases Google Cloud names are a tell: rapid idea generation, A/B testing ad variations, and social apps serving millions of users. Those are not single-player creative toys. They are high-volume workflows where regeneration cost becomes product gravity. This is the pricing page as a grocery receipt: each line looks manageable until the cart is full.

Lite is a lane, not a demotion Ars Technica reports that Nano Banana 2

Lite is part of the Gemini 3.1 family and is technically called Gemini 3.1 Flash Lite Image. Ars also notes Google’s positioning around exploring ideas and rapid prototyping, where quality can sometimes take a backseat to speed. That is exactly how a Lite model should be judged. If the job is concept exploration, ad variation, or user-generated creative flow, then a cheaper and faster first pass can be more valuable than a slower premium render. Google’s Keyword post, dated Jun 30, 2026, frames Nano Banana 2 Lite as the company’s fastest and most cost-efficient Gemini Image model. The practical lesson for builders is to stop treating media generation as one undifferentiated feature. A healthy product architecture will route cheap drafts, iterative edits, and final assets differently. Otherwise, your gross margin becomes a Choose Your Own Adventure where every ending is expensive.

Gemini Omni Flash makes video feel like a workflow Google Cloud says Gemini Omni

Flash is grounded in Gemini’s real-world knowledge and supports high-quality video generation and conversational editing. Google’s earlier Gemini Omni post describes the broader Gemini Omni idea as allowing users to create from any input and edit naturally using conversational language. The important product shift is not simply text-to-video. It is video creation as an interactive loop, where the user asks for a change rather than reopening the entire production process. That preview status is worth respecting. Public preview means builders can start testing workflow fit, but should be careful about committing customer-facing economics before pricing, latency, quota behavior, and reliability are locked into their own operating model. The cited launch snippets do not disclose a per-second output price for Gemini Omni Flash, so margin planning should include a verification step before any public SKU or seat-based bundle depends on it. The smart move is to prototype the interaction model now and hold back hard promises until the cost envelope is fully visible.

The next move is routing, not razzle-dazzle

The competitive map Google is drawing, based on Google Cloud’s launch language, is a tiered media stack: lower-cost image iteration at scale, plus conversational video editing moving into preview. That gives developers a more realistic way to design product tiers. Free users might get fast draft images, paid users might get higher limits or video workflows, and enterprise customers might pay for governance, reliability, and integration through the Gemini Enterprise Agent Platform. For startup teams, the takeaway is simple: model the regeneration loop before you ship the prompt box. If users love the feature, they will use it more than your demo spreadsheet assumed. Watch next for clearer Gemini Omni Flash pricing, quota details, and whether Google makes routing across image and video models easier for developers. The winning AI media products will not be the ones with the flashiest launch clip. They will be the ones whose unit economics still look sane after users show up.