nano-banana-image-gen

To generate images from text prompts using Gemini models, you can use the Nano Banana API on Orthogonal, which supports both flash for quick drafts and pro for high-quality outputs across diverse styles. It routes your prompt through Gemini models to produce the visual.

Yes, generating images with Nano Banana on Orthogonal requires a Gemini API key with active credits, alongside the Orthogonal CLI configured with proper authentication. These prerequisites ensure your text prompts are processed successfully by the Gemini models.

The difference between flash and pro quality tiers in Gemini image generation is output fidelity. The flash tier uses gemini-2.5-flash-image for quick drafts and concept visuals, while the pro tier uses gemini-3-pro-image-preview for high-quality outputs and detailed creative mockups.

Yes, you can use the Orthogonal CLI for automated text-to-image workflows. The Nano Banana API integrates directly with the CLI, allowing you to script image generation from text prompts for streamlined design exploration and predictable creative mockups.

Limitations of generating images via the Nano Banana API include dependency on having a Gemini API key with sufficient credits and a properly authenticated Orthogonal CLI environment. Without these, the flash and pro image generation processes cannot execute.