← Back to Catalog
Agent Skill

nano-banana

Generate images with Google Gemini. Fast, cheap, good for iteration.

Generate and edit images using Google's Gemini image generation models (Nano Banana family). Supports style presets, platform-specific sizing (YouTube/slides/blog), variants, image editing via inlineData, reference images for style transfer, and organized output with metadata. Default model is Nano Banana 2 (gemini-3.1-flash-image-preview). Key is auto-decrypted via SOPS.

What it does

Generates and edits images from text prompts using Google’s Gemini image generation models. Fast, cheap, and good for rapid iteration. Three model tiers from quick flash to highest-quality pro.

Models

  • Nano Banana 2 (default) — gemini-3.1-flash-image-preview, best instruction following, fast
  • Nano Banana Pro — gemini-3-pro-image-preview, highest quality, best text in images
  • Nano Banana (original) — gemini-2.5-flash-image, legacy

Key features

  • Style presets — editorial (thin lines, muted palette), wireframe, grain, and more
  • Platform sizing — auto-crop for YouTube thumbnails, slides, blog headers
  • Variants — generate N versions with a contact sheet for comparison
  • Image editing — modify existing images with text instructions
  • Style transfer — match the aesthetic of a reference image
  • History — re-roll the last prompt or browse generation history

When to use

When you need quick image generation for presentations, thumbnails, articles, or social posts. Faster and cheaper than gpt-image-2; choose that one instead when text rendering quality is critical.