Google Gemini now uses photos to understand your taste and create custom images

Google has launched a new update to its Gemini app called Nano Banana 2 for “personal intelligence” with its image-generating model. With this update, Gemini can now pull context directly from services like Google Photos to create images that truly reflect your real life. You no longer need to upload reference images or explain everything in detail. Simple tips like “Design my dream home” can now generate results based on your taste and lifestyle.

For the past seven years, I’ve tracked consumer technology through the constant changes in hardware, platforms, and the way people actually use their devices. Covering everything from budget gear to flagship hardware, I focus on what readers need to know, not buzzwords or release cycle hype. My expertise spans gaming laptops and gaming chairs, high-performance computers, gaming monitors, printers, smartwatches, headphones, headsets, Bluetooth speakers, tablets, and more, with a special focus on how these products perform in daily use. Reviews, explainers, buying guides, and news stories all have a common goal: to provide readers with enough detail to make a confident decision rather than wading through nonsense. Outside of deadlines, I spend a lot of time playing games, watching movies, and anime, which naturally flows into work. Performance, comfort, display quality and sound are judged on how players and viewers experience it, not just lab data, which grounds my coverage in real-life scenarios, not just benchmarks.

Read moreRead less

It’s powered by Nano Banana 2, Google’s latest imaging model, which works with personal intelligence to fill in the gaps using data from your Google account. The goal is to reduce the amount of work required for prompts while making the final output feel more relevant and useful.

Another standout feature is the ability to include real people in the generated images. By linking Google Photos, Gemini can identify tagged photos of friends and family and use them as reference. You can also adjust the results, swap images or try different styles such as watercolor or claymation.

Behind the scenes, the system also uses metadata such as photo tags and activity context to identify people and preferences, helping the model generate more accurate and consistent visuals while maintaining high-quality output at faster image generation speeds.

Google says privacy remains a top priority here. Your personal photos are not used to train the AI ​​model, and the feature is entirely opt-in.

The service has begun rolling out to Gemini AI Plus, Pro and Ultra users in select regions, with a wider release expected soon. The move signals Google’s push toward more context-aware AI tools that create output that feels less generic and more personal.

The update also makes everyday creative tasks faster, especially for casual users who want quick results without learning tips and tricks. It lowers the barrier to entry and makes advanced image generation easier and more practical.