Google opens Nano Banana 2 Lite and Omni Flash to developers

incrementalDeveloperPlatform

Wednesday, July 1, 2026

Confidence

High

Evidence

Google Cloud blog + Google DeepMind blog + developer trade press

Google made its fastest image model generally available and moved its Gemini video model into public API preview.

The details

Google announced general availability of Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image), released Gemini Omni Flash in public preview, and added both to the Gemini Enterprise Agent Platform. Google Cloud
Nano Banana 2 Lite generates images in about 4 seconds starting at $0.034 per 1,000 images; Omni Flash is priced at $0.10 per second of video output, the same rate as Veo 3.1 Fast. Techgenyz
Google is transparent about preview limits: video generations are capped at 10 seconds, audio references aren't supported, scene extension isn't in the API, video references up to 3 seconds are accepted but not processed correctly, and character consistency across scene changes is still limited.

What it means

Two efficiency-tier releases on the same day the Anthropic story dominates every headline; the timing is not accidental. Google's argument is that the workflow is what matters — chain Nano Banana 2 Lite into Omni Flash and you have a text-to-image-to-video pipeline inside one vendor's SDK at commodity pricing. Developers building agentic creative apps get a cheaper substitute for stacked third-party services. Enterprise buyers wanting Veo-class video without the Veo cost now have a preview to run against real workloads before Q3 planning locks.

Google made its fastest image model generally available and moved its Gemini video model into public API preview.

Google announced general availability of Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image), released Gemini Omni Flash in public preview, and added both to the Gemini Enterprise Agent Platform. Google Cloud
Nano Banana 2 Lite generates images in about 4 seconds starting at $0.034 per 1,000 images; Omni Flash is priced at $0.10 per second of video output, the same rate as Veo 3.1 Fast. Techgenyz
Google is transparent about preview limits: video generations are capped at 10 seconds, audio references aren't supported, scene extension isn't in the API, video references up to 3 seconds are accepted but not processed correctly, and character consistency across scene changes is still limited.

Sources

Google opens Nano Banana 2 Lite and Omni Flash to developers

Turn a Payer Policy-Update Bulletin into Revenue-Cycle Action Items

OpenAI GPT-5.6 (Sol, Terra, Luna preview)