Google’s Nano Banana 2 fixes blurry text and boosts speed — here's everything included in this massive upgrade

Fans of Nano Banana’s original release have something new to get excited about. Today, Google DeepMind unveiled Nano Banana 2 (officially Gemini 3.1 Flash Image), a next-generation model that blends the high-fidelity intelligence of the Pro version with the lightning-fast performance of Gemini Flash.

The promise is professional-grade images with quicker and smoother results, including a much-needed upgrade to on-image text.

Here's a deeper dive into what's new.

Nano Banana 2 — Pro quality with Flash speed

The new model introduces several advanced capabilities aimed at giving users ultimate creative control and real-world accuracy:

• Real-time web grounding: Nano Banana 2 draws on Gemini’s advanced world knowledge and live web search data to render specific subjects more accurately.
• Next-level text rendering: The model can generate perfectly legible text inside images for marketing mockups or greeting cards. It can even translate and localize text directly within a generated image so you can share ideas globally.
• Data visualization: Thanks to its deep understanding, the model can generate infographics, turn notes into diagrams and create accurate data visualizations.
• Creative control and consistency Google says Nano Banana 2 significantly narrows the gap between speed and realism. Key improvements include: subject consistency, better instruction following, production-ready output and higher visual fidelity.

Nano Banana 2 promises to support your creative process be it professionally or personally. You can keep the exact look of up to five characters and 14 objects consistent across a single workflow, which makes storyboarding or building visual narratives dramatically easier.

It’s also better at following complex, layered prompts, so the image you get is much closer to what you actually envisioned. With production-ready outputs and native aspect ratios and resolutions ranging from 512px all the way up to crisp 4K, the results are shockingly better than the first generation Nano Generation.

Availability

Nano Banana 2 is rolling out immediately as the default image generator across Google’s ecosystem. In the Gemini app, it replaces Nano Banana Pro across Fast, Thinking and Pro modes, though Google AI Pro and Ultra subscribers can still access the Pro model for specialized tasks through a menu option.

It’s also arriving in Google Search via AI Mode and Lens across 141 countries, while developers can access it through Google AI Studio and the Gemini API. For enterprise use, it’s available in Vertex AI on Google Cloud. In Flow, Nano Banana 2 becomes the default image model at zero credits, and it will also power creative suggestions inside Google Ads.

Bottom line

Starting today, Nano Banana 2 underscores the possibilities of AI image generation. It's clear that multimodal capabilities are maturing into a trustworthy, everyday tool, especially with built-in SynthID watermarking and C2PA credentials, to verify how content was created.

More importantly, the model promises to close the gap between speed and quality, so you're spending less time waiting for image generation and ultimately getting something genuinely practical for real work — from marketing visuals to storytelling and rapid creative iteration. Give it a try and let me know in the comments what you think.

Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds.

More from Tom's Guide

• I use the 'reverse brief' prompt to instantly understand any document — here’s how it works
• I use the 'hype check' prompt before every big decision — it's stopped me making 3 huge mistakes
• Gemini 3.1 Pro is a powerhouse for deep work — here are 7 prompts that prove it