Google has introduced Nano Banana 2, officially branded as Gemini 3.1 Flash Image, expanding its image generation portfolio with a model designed to combine advanced creative intelligence and rapid performance. The new release blends the visual fidelity and reasoning capabilities of Nano Banana Pro with the speed of Gemini Flash, positioning it as a production-ready tool for a wide range of creative workflows.
Nano Banana 2 builds on the momentum of the original Nano Banana and the later Pro variant, bringing high-level features into a faster, more broadly accessible model across Google’s ecosystem.
Intelligence and Visual Quality at Flash Speed
Nano Banana 2 integrates Gemini Flash’s processing speed into image generation, enabling faster edits, rapid iteration, and near real-time visual experimentation. For designers, marketers, and content creators, that translates into shorter turnaround times without sacrificing quality.
The model incorporates advanced world knowledge drawn from Gemini’s broader intelligence stack. It leverages real-time information and images from web search to render specific subjects more accurately. That capability extends beyond photorealistic scenes. Users can generate infographics, convert notes into diagrams, and produce data visualizations with contextual accuracy.
Text rendering receives a meaningful upgrade as well. Nano Banana 2 can generate legible, precise text inside images, making it suitable for marketing mockups, greeting cards, and social content. It can translate and localize text directly within visuals, supporting global creative campaigns without requiring external editing tools.
Enhanced Creative Control and Subject Consistency
One of the more technical upgrades in Nano Banana 2 is improved subject consistency. The model can maintain character resemblance for up to five characters and preserve the fidelity of up to 14 objects within a single workflow. For storyboarding, multi-scene narratives, or product visualization sequences, that consistency removes much of the friction creators often encounter when refining iterations.
Instruction following has been strengthened, with the model adhering more strictly to detailed prompts. Complex requests involving multiple elements, stylistic constraints, or nuanced scene composition are handled with tighter alignment between prompt and output.
Production-ready specifications are another highlight. Users can select aspect ratios and resolutions ranging from 512 pixels up to 4K. That flexibility makes the model adaptable for everything from vertical social posts to widescreen presentations or digital signage.
Google says Nano Banana 2 closes the gap between speed and visual fidelity. The model delivers sharper details, richer textures, and improved lighting, all within the performance expectations of the Flash tier.
Availability Across Google Products
Nano Banana 2 is rolling out across several Google platforms starting today.
Within the Gemini app, Nano Banana 2 replaces Nano Banana Pro across the Fast, Thinking, and Pro model options. Google AI Pro and Ultra subscribers retain access to Nano Banana Pro for specialized, high-fidelity tasks through a regeneration option in the interface.
In Search, Nano Banana 2 is available through AI Mode and Lens in the Google app, as well as mobile and desktop browsers. Google notes expanded availability across 141 additional countries and territories, along with support for eight more languages.
Developers can access Nano Banana 2 in preview via AI Studio and the Gemini API. It is available in Google Cloud through Vertex AI in preview as part of the Gemini API. The model is now the default image generation engine in Flow, where it is accessible to all users without consuming credits.
Google Ads integrates Nano Banana 2 to power creative suggestions during campaign creation, giving advertisers faster visual asset generation directly within the platform.
Choosing Between Nano Banana 2 and Pro
Google now positions its lineup as purpose-driven. Nano Banana Pro remains available for workflows demanding maximum factual accuracy and high-fidelity output. Nano Banana 2 focuses on rapid generation, strict instruction adherence, and integrated grounding through image search.
For teams balancing speed and creative precision, Nano Banana 2 is framed as the everyday workhorse. More details about the rollout and platform availability can be found on Google’s official announcement page.
