skills.rest
Aadi-110i
Aadi-110iCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

nano-banana-2

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

Updated Jul 28, 2026
Search Tags:#cli#image generation#image editing#gemini#google#text-to-image#inference.sh
One-click install
npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill nano-banana-2-aadi-110i
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nano-banana-2
Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/nano-banana-2
Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill nano-banana-2-aadi-110i

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating or editing images programmatically often requires managing API keys, SDKs, and model endpoints. This Skill provides direct command-line access to Google's Gemini 3.1 Flash Image Preview model through the inference.sh CLI, covering text-to-image generation, image editing, and multi-image inputs without custom integration code. ## Core Features & Use Cases - Text-to-Image Generation: Create images from text prompts with control over aspect ratio (1:1, 16:9, 9:16, 4:3, 3:4), resolution (1K, 2K, 4K), and number of output images. - Image Editing with Multi-Image Input: Supply up to 14 input images (JPEG, PNG, WebP) and describe edits in natural language, such as adding elements or changing scenes. - Google Search Grounding: Enable real-time information grounding for prompts that depend on current data like weather or news. - Use Case: A content creator needs four logo variations for a coffee shop. They run a single belt command with a prompt and num_images set to 4, then iterate on the best result with an editing prompt. ## Quick Start Ask the AI to generate an image with Gemini 3.1 Flash Image Preview, for example: "Generate a photorealistic image of a banana in space using nano banana 2."

Frequently Asked Questions about nano-banana-2

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate images with Gemini 3.1 Flash Image Preview from the command line?▼

Run belt app run google/gemini-3-1-flash-image-preview with a JSON input containing your prompt. You must first install the inference.sh belt CLI and authenticate with belt login.

How to edit an existing image with Nano Banana 2?▼

Pass an images array with up to 14 image URLs (JPEG, PNG, or WebP) alongside a prompt describing the change, such as adding a rainbow to the sky. The model returns the edited images in the output.

What aspect ratios and resolutions does Gemini 3.1 Flash Image support?▼

Supported aspect ratios are 1:1, 16:9, 9:16, 4:3, 3:4, and auto. Resolutions include 1K, 2K, and 4K, with 1K as the default.

Can Gemini 3.1 Flash Image use real-time information in prompts?▼

Yes, set enable_google_search to true in the input JSON. This grounds the generation in real-time data such as current weather or news.

Is there a Python SDK alternative to the belt CLI for image generation?▼

Yes, the inferencesh Python package provides a client.run method accepting the app identifier and input dictionary. It also supports streaming progress updates during generation.

Related Skills

Explore More Design & Creative Skills→
WebWakaHubWebWakaHub
20

Image Prompt Engineer

Craft detailed photography prompts for AI image generation tools.

Official
Advanced
NEXUZ-SYSNEXUZ-SYS

image-fetcher

Acquire web images, Playwright screenshots, and user files into a reference folder.

Official
Intermediate
cloudofgeorgecloudofgeorge

benchmark-models

Compare AI models by latency, cost, and output quality.

Community
Advanced
marmikcfcmarmikcfc

nano-banana-image-gen

Generate images from text prompts via the Nano Banana API on Orthogonal.

Community
Intermediate