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

nano-banana

Generate and edit images with Google Gemini native image models 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-aadi-110i
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nano-banana
Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/nano-banana
Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill nano-banana-aadi-110i

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating or editing images with Google's Gemini native image models requires API setup and integration work; this Skill lets you generate images directly from the command line through the inference.sh platform with a single command. ## Core Features & Use Cases - Text-to-Image Generation: Create images from text prompts using Gemini 3 Pro Image (best quality) or Gemini 2.5 Flash Image (faster). - Image Editing: Supply up to 14 input images and modify them with natural language instructions. - Output Control: Configure aspect ratio, resolution up to 4K, number of images, output format, and Google Search grounding. - Use Case: A designer needs four logo variations for a coffee shop; run one command with a prompt and num_images set to 4 to receive multiple options instantly. ## Quick Start Ask the AI to generate an image of a banana in space using the Gemini 3 Pro Image model via the belt CLI.

Frequently Asked Questions about nano-banana

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

FAQPage Schema
How do I generate images with Gemini from the command line?▼

Install the inference.sh belt CLI, run belt login, then execute belt app run google/gemini-3-pro-image-preview with a JSON input containing your prompt. The generated image is returned by the CLI.

What is the difference between Gemini 3 Pro Image and Gemini 2.5 Flash Image?▼

Gemini 3 Pro Image is slower but produces the best quality and supports up to 4K resolution. Gemini 2.5 Flash Image is faster with excellent quality, suited for quick iterations.

Can Gemini image models edit existing images?▼

Yes, pass an images array with up to 14 input image URLs along with a prompt describing the change, such as adding elements or altering the scene. Both Gemini image models support this editing mode.

Does Gemini image generation support custom aspect ratios and resolution?▼

Yes, set aspect_ratio to values like 1:1, 16:9, 9:16, 4:3, or 3:4, and resolution to 1K, 2K, or 4K. Note that 4K resolution is only available on Gemini 3 Pro Image.

What do I need before running the nano banana image commands?▼

You need the inference.sh belt CLI installed and authenticated via belt login. Without an authenticated CLI session, the belt app run commands cannot execute.

Related Skills

Explore More Design & Creative Skills→
inference-shinference-sh
723

gpt-image

Generate and edit images with OpenAI GPT-Image-2 via the inference.sh CLI.

Official
Basic
yangyixxxxyangyixxxx
13

gptnb-image

Generate images, layered assets, and videos via the gptnb.ai middleware API.

Community
Advanced
j-chingj-ching
1

nano-banana-2

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

Community
Intermediate
docaohieu2808docaohieu2808

nano-banana-2

Generate and edit images via Google Gemini 3.1 Flash using the inference.sh CLI.

Community
Intermediate