Nano Banana 2

This wraps Google's Gemini 3.1 Flash Image Preview model through the inference.sh CLI, giving you text-to-image generation, image editing with up to 14 input images, and Google Search grounding for real-time info. It's the sequel to the original nano-banana skill, updated for the newer Gemini model. You get options for aspect ratios, resolution up to 4K, and multi-image outputs. The grounding feature is interesting if you want current weather or news data baked into generated images. Requires the belt CLI and an inference.sh account, but once set up it's just bash commands or Python SDK calls. Good pick if you're already in the inference.sh ecosystem and want quick access to Google's image models without dealing with their API directly.

skills.sh installs: 1.1K
GitHub stars: 804
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add inference-sh/skills --skill nano-banana-2 --agent claude-code

Installs into .claude/skills of the current project.

Files

Install the belt CLI skill: npx skills add belt-sh/cli

Nano Banana 2 - Gemini 3.1 Flash Image Preview

Generate images with Google Gemini 3.1 Flash Image Preview via inference.sh CLI.

Quick Start

Requires inference.sh CLI (belt). Install instructions

belt login

belt app run google/gemini-3-1-flash-image --input '{"prompt": "a banana in space, photorealistic"}'

Examples

Basic Text-to-Image

belt app run google/gemini-3-1-flash-image --input '{
  "prompt": "A futuristic cityscape at sunset with flying cars"
}'

Multiple Images

belt app run google/gemini-3-1-flash-image --input '{
  "prompt": "Minimalist logo design for a coffee shop",
  "num_images": 4
}'

Custom Aspect Ratio

belt app run google/gemini-3-1-flash-image --input '{
  "prompt": "Panoramic mountain landscape with northern lights",
  "aspect_ratio": "16:9"
}'

Image Editing (with input images)

SKILL.md on GitHub

Details

Repository: github.com/inference-sh/skills
Categories: CLI & Terminal, Generative Media
First seen: Jun 3, 2026