What you need to know
- Nano Banana 2 brings faster image generation with higher precision and improved text rendering.
- The new model now supports 512px to 4K image outputs in multiple aspect ratios.
- Google is making Nano Banana 2 image generation available to all Gemini users, not just paid subscribers.
- Nano Banana 2 is now the default model across Gemini, Search AI mode, Lens, and developer tools.
Google has just taken the wraps off its new image generation model called Nano Banana 2, and it’s a major upgrade over the previous version.
One of the most common use cases of AI chatbots and models today is image generation. People love turning text prompts into images, and Google has been pushing hard in this space. The company introduced Nano Banana in the second half of 2025 and later upgraded it to Nano Banana Pro in November, which went viral. Now Google has officially unveiled Nano Banana 2.
Nano Banana 2, technically called Gemini 3.1 Flash Image, brings everything good about Nano Banana Pro with higher precision and faster performance. Google says Nano Banana 2 has access to real-time knowledge from the web and improved text rendering, allowing it to generate accurate and legible text for marketing mockups or greeting cards.
Image 1 of 3
The biggest change this time is availability. Nano Banana 2 brings Pro-level image generation to all Gemini users, not just paid subscribers. It can generate images ranging from 512 pixels all the way up to 4K in multiple aspect ratios without issues. The model also follows complex prompts more strictly, so you get closer to exactly what you asked for.
Consistency has always been one of the biggest challenges in image generation, but Google claims Nano Banana 2 maintains character resemblance for up to five characters and object fidelity for up to 14 objects within a single workflow. It also promises more vibrant lighting, richer textures, and sharper detail overall.
Nano Banana 2 now replaces Nano Banana Pro for most users in the Gemini app. However, Google AI Pro and Ultra subscribers will still have access to Nano Banana Pro through the three-dot menu. Google is also making it easier to generate images by letting users choose specific visual styles for their outputs.
Nano Banana 2 is becoming Google’s default image generation model across platforms. It is available in Gemini, Google Search’s AI mode, Google Lens, AI Studio, Gemini API, Gemini CLI, and more. The rollout has already begun, and users can start accessing it today.
Android Central’s Take
I haven’t used Nano Banana 2 yet, but considering how much better Nano Banana Pro was compared to other AI image generation tools, I’m already optimistic. Google has been steadily improving its image models, and I wouldn’t be surprised if it has pulled off another solid upgrade this time.
Thankfully, the company is also taking safety around AI-generated images more seriously. Google has now coupled its own SynthID watermarking technology with interoperable C2PA Content Credentials, which are already used by companies like Meta, OpenAI, Adobe, and others.
Google says it’ll now provide users with a clearer and more contextual view of not just whether AI was used to generate an image, but also how it was used.
