Google’s Gemini AI has expanded its image generation capabilities in the United States, allowing users to create and edit images based on their own photos through the Gemini Apps platform. This advancement leverages the Nano Banana 2 model, which Google introduced as its latest multimodal image generation system, enabling users to transform personal photos into modern visual concepts using natural language prompts.
The feature, now available to Gemini Apps users in the U.S., represents a significant step in making advanced AI image editing accessible to consumers. With Nano Banana 2, users can upload their own photographs and instruct the AI to modify elements such as lighting, background, clothing, or artistic style—all through simple text commands. For example, a user could take a casual selfie and generate a version where they appear in a futuristic cityscape during golden hour, or convert a color photograph into a black-and-white portrait with adjusted focus and texture.
According to Google’s official support documentation for Gemini Apps, Nano Banana 2 offers improved world knowledge, character consistency and localized editing capabilities. The model can maintain the identity of a person or object across multiple generated images, making it useful for creating cohesive visual narratives or product visualizations. Users can also combine multiple reference images into a single output, enabling creative scene composition without requiring professional design skills.
Access to these features requires users to be signed into their Gemini Apps account, with certain restrictions applying based on account type and age. Individuals under 18 years old (or the applicable age in their country) using personal Google accounts cannot access image generation. Work or school accounts may have additional limitations depending on institutional licensing agreements. Google emphasizes that users must comply with its Terms of Service and Prohibited Use Policy when generating content, including respecting copyright and privacy rights.
The underlying technology powering these capabilities is part of Google’s broader Gemini 2.5 Flash Image model family, also referred to as ‘nano-banana’ in developer and user communities. This multimodal system understands both textual and visual inputs, allowing it to interpret complex prompts involving scenes, actions, and stylistic transformations. Google positions the tool as beneficial for work, personal projects, and creative experimentation, noting that paying subscribers gain access to Nano Banana Pro for image regeneration.
While the ability to generate images from personal photos opens new creative possibilities, Google advises caution regarding the use of AI-generated content. The company recommends that users critically evaluate outputs before trusting, publishing, or using them, particularly in contexts where accuracy or authenticity is essential. As with all generative AI systems, outputs may reflect biases or inaccuracies present in training data, and users bear responsibility for how they employ the technology.
This rollout aligns with Google’s ongoing effort to integrate advanced AI features into its consumer applications while maintaining safeguards around misuse. The image generation functionality in Gemini Apps builds upon earlier versions of the model, incorporating feedback to improve reliability, speed, and output quality. Users interested in trying the feature can access it through the Gemini Apps interface by selecting the image creation tool and choosing from available model options such as Speedy, Think, or Pro.
As AI-powered image editing becomes more widespread, Google’s approach with Gemini Apps highlights a balance between innovation and user responsibility. By providing powerful creative tools alongside clear usage guidelines, the company aims to empower individuals to explore AI-assisted design while mitigating risks related to misuse, misinformation, or intellectual property violations.
For the latest updates on Gemini Apps features and availability, users are encouraged to consult Google’s official support pages and in-app notifications, which provide region-specific information about functionality and access requirements.
Related reading