Google's Nano Banana AI Image Model Upgraded to Pro—Free Trial Available

Google has just unveiled Nano Banana Pro (Gemini 3 Pro Image), an enhanced image generation and editing model built on Gemini 3 Pro. The company claims it can "transform your ideas into studio-quality designs with unprecedented control, flawless text rendering, and enriched world knowledge." This marks a significant leap from the hyper-realistic 3D doll images that propelled the original model to viral fame back in September.

Starting today, users worldwide can access this capability for free within the Gemini app by selecting “Create Image” and using the “Think” model.

According to Google, Nano Banana Pro excels at generating context-rich infographics and charts that visualize real-time data—such as weather updates or sports statistics. It renders legible text directly onto images, whether short slogans or full paragraphs, making it ideal for creating multilingual posters or invitations. The model also supports compositing up to 14 source images and incorporating up to five people into a single cohesive scene.

Thanks to its advanced inpainting capabilities, Nano Banana Pro introduces powerful new creative controls. Users can selectively edit any part of an image locally—adjust camera angles, apply bokeh effects, shift focus, fine-tune color grading, or even transition lighting from daytime to nighttime. The model supports outputs up to 4K resolution and accommodates multiple aspect ratios.

Importantly, all images created or edited with Nano Banana Pro will include C2PA metadata. Once widely adopted across platforms, this should make it easier to identify AI-generated content or deepfakes in search results and social media feeds. TikTok announced this week that it will also use C2PA metadata to embed invisible watermarks on AI-generated content.

Free-tier users of Nano Banana Pro will be subject to usage quotas, which are expanded for subscribers of Google AI Plus, Pro, and Ultra plans. Additionally, Google AI Pro and Ultra subscribers in the U.S. can access AI Mode in Search, while NotebookLM research assistant users gain global access to these features.