Build with Nano Banana Pro, our Gemini 3 Pro Image model
Here’s how developers can get started with Nano Banana Pro (Gemini 3 Pro Image) — an incredibly powerful new image creation and editing model that gives you more control, more detail, and way more creative freedom.
Today, we’re introducing Nano Banana Pro (Gemini 3 Pro Image), a higher-quality image model built on top of Gemini 3 Pro, giving developers access to stunning, studio-level visuals. It feels exciting to finally share this, especially after releasing Nano Banana (Gemini 2.5 Flash Image) just a few months back. Since then, it’s been genuinely heartwarming to watch the community explore its features — from keeping characters consistent, to bringing old photos back to life, and even making tiny edits across a massive endless canvas.
This next-generation image model is beginning its paid preview rollout, opening the door to a new wave of smart, multimodal apps built with the Gemini API in Google AI Studio and in Vertex AI for enterprise teams. It delivers extremely high-fidelity images, sharper and more accurate text rendering, and strong world knowledge — all boosted by the model’s ability to ground responses with Google Search and pull in useful information directly from your prompt.
 |
| Gemini 3 Pro Image excels on Text to Image AI benchmarks. |
We’re also broadening where Gemini 3 Pro Image can be used across the developer world. In Google Antigravity — our new agent-driven development platform — coding agents can tap into these image tools to instantly create detailed UI mockups for user feedback, or generate brand-new visual assets before writing any code. Major creative platforms are also jumping on board, including Adobe and Figma, bringing these capabilities right into the tools people already love.
High fidelity and controls
If you’re creating something detailed and precise, Gemini 3 Pro Image gives you full control over things like lighting, camera angles, focus, and color tones. It feels empowering to shape every part of the image so it looks exactly the way you imagined—clean, sharp, and truly professional.
 |
A silhouette lost in a sea of golden bokeh and morning mist.
Prompt: Replace volumetric lighting with bokeh |
With both 2K and 4K resolution options, you can relax knowing your final images will meet high-quality standards for real production work. You can easily build ads that look smooth and consistent by mixing different pieces together—product photos, logos, reference images, anything you need. Keep the same realistic look for up to five people, pull in six high-detail shots, or blend up to fourteen regular images into one polished, finished ad. And if you want to play around creatively, try our demo app—it lets you match logos with products and design your own mockups in seconds.
Improved text rendering and localization
Gemini 3 Pro Image brings a huge upgrade compared to 2.5 Flash Image. It turns ideas that used to feel vague or abstract into clear, useful visuals. The model handles both logic and language with surprising accuracy, and its text rendering feels almost magical — crisp, readable, and naturally blended into your images.
 |
Creative food photography where each word is artistically spelled out using the actual ingredients associated with that food.
Prompt: Make 8 sophisticated minimalistic logos, each is a fun food word, and make letters from realistic food to express the meaning of this word. composition: a rendering of all logos on a single solid white background |
It’s perfect for creating marketing materials, educational content, and so many other real-world projects. You can even try it out using our comic book generator in Google AI Studio. It’s genuinely fun — you can make your own multi-page comic starring you and a friend, complete with beautifully styled text and rich visuals.
With Gemini 3 Pro Image, we’ve finally erased the gap between image creation and smart language adaptation. The model actually understands the meaning of what’s in an image, so switching languages on things like signs, menus, or documents becomes incredibly smooth. And thanks to image-to-image generation, you can change the text while keeping the original art style or layout intact.
 |
A beverage campaign concept showcasing accurate translation and rendering of English text into French.
Prompt: Translate to French |
Access to the world’s knowledge
Gemini 3 Pro Image taps into an enormous library of information, helping it create visuals that feel more accurate and trustworthy than older image models. And when you turn on Google Search grounding, it can even pull in fresh, real-time details from the web. This makes a huge difference for things that need to be spot-on—like science diagrams, historical maps, or anything where the facts really matter. You can see the magic yourself using our demo app, where you can instantly build customized infographics on any topic, crafted just the way your audience needs.
 |
| Bike care and maintenance infographic generated from a demo app that creates educational infographics. |
Go wild and start creating today
This new model update includes so much of the feedback you’ve already shared with us — and honestly, we’re excited, because we’re just getting started. To help everyone trust where AI-generated images come from, we’ve added SynthID digital watermarks to every picture made or edited with Gemini 3 Pro Image. It’s our way of keeping things transparent and honest.
 |
| Use Gemini 2.5 Flash Image for faster, lower cost image generation, or 3 Pro Image for higher quality image generation, with higher cost and latency. |
Take a moment to explore the apps built with Gemini 3 Pro Image — they’re fun, inspiring, and might spark that creative spark you’ve been waiting for. And once you’re feeling that excitement (we hope you do!), try remixing these demo apps or plug the model directly into your own projects using the Gemini API in Google AI Studio or Vertex AI if you’re building for enterprise.
If you ever feel stuck or curious along the way, dive into the documentation, prompt guide, or cookbook — or visit the developer forum to get help, share ideas, and connect with others who are building right alongside you.
