nano-banana-2

The nano-banana-2 skill leverages Google Gemini 3.1 Flash Image Preview to generate and edit images through the inference.sh CLI. It provides a streamlined interface for creating visuals from textual prompts or modifying existing images, making it ideal for designers, content creators, and developers who need quick, high-quality imagery without manual graphic design. By integrating with the command-line workflow, users can automate image generation tasks efficiently while maintaining control over output parameters and quality.

Key features include text-to-image generation, image editing with up to 14 input images, customizable aspect ratios, high-resolution outputs up to 4K, and Google Search grounding for real-time contextual content. The skill supports various image formats such as JPEG, PNG, and WebP, and offers advanced prompt customization, allowing users to specify styles, lighting, composition, and detailed visual elements. Additionally, it integrates with Python SDKs, enabling programmatic access and live streaming of generation progress.

Nano-banana-2 is particularly useful for digital artists, marketing teams, UX/UI designers, and developers who require rapid prototyping of visuals or automated image creation for web and mobile applications. It can be used to create promotional graphics, concept art, UI mockups, social media content, and context-aware illustrations. Its combination of versatility, high-quality output, and automation capabilities makes it suitable for both individual creatives and teams in professional settings.