skills.rest
szweibel
szweibelCOMMUNITY
Explore all skills by this author →
GitHub

nano-banana-pro

Generate 4K images from text prompts via the Gemini 3 Pro Image API.

2|2|Updated Nov 6, 2025
Search Tags:#grounding#gemini#image-generation#prompting#4k
One-click install
npx skills add https://github.com/szweibel/claude-skills --skill nano-banana-pro-szweibel
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nano-banana-pro
Source: https://github.com/szweibel/claude-skills/tree/main/nano-banana-pro
Command: npx skills add https://github.com/szweibel/claude-skills --skill nano-banana-pro-szweibel

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, Pillow, and includes assets (resource) and references (resource) and scripts (resource) components.

What problem does it solve?

Facilitates access to Google's Nano Banana Pro (Gemini 3 Pro Image) API for high-quality image generation with 4K support, prompt best practices, and real-time collaboration features.

Core Features & Use Cases

  • API usage & prompting guide: documented prompts and code examples.
  • 4K generation & ground truth: high-resolution images suitable for print and posters.
  • Multi-turn iteration: progressive refinement through multiple prompts.
  • Use Case: generate a product concept image in 4K, iteratively refine colors and typography through multiple prompts.

Quick Start

Set up API key, install dependencies, and run example_basic.py to generate a simple image.

Frequently Asked Questions about nano-banana-pro

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

FAQPage Schema
How do I generate 4K images from text prompts using Google's Gemini API?▼

Use the Gemini 3 Pro Image API with the gemini-3-pro-image-preview model and 4K image_config to generate high-resolution images. This Skill provides Python and JavaScript examples covering API key setup, dependency installation, and prompt best practices for production-quality output.

What prompting techniques produce better results with Gemini image generation?▼

Effective image prompts combine clear subject descriptions, visual style specifications, and technical parameters. This Skill documents prompting best practices and includes iterative refinement workflows to progressively improve generated images through multiple turns.

Can I refine generated images iteratively with the Gemini API?▼

Yes, the Gemini 3 Pro Image API supports multi-turn refinement. You can generate an initial image, then submit follow-up prompts to adjust colors, typography, composition, and other details across multiple iterations in a single workflow.

Do I need search grounding to generate images with Gemini?▼

Search grounding is optional and enhances results when you want generated images grounded in real-world references. This Skill covers both search-grounded and non-grounded workflows with multi-image reference support for flexible use cases.

What are the dependencies and setup requirements for Gemini image generation?▼

You need google-genai for API access and Pillow for image processing. The Skill provides step-by-step setup including API key configuration, dependency installation via package managers, and client initialization for immediate use.

Can I use Gemini image generation with both Python and JavaScript?▼

Yes, this Skill includes working examples in both Python and JavaScript. Both implementations handle API authentication, 4K image configuration, TEXT and IMAGE response modalities, and optional search grounding identically.

Related Skills

Explore More Design & Creative Skills→
drshailesh88drshailesh88

nano-banana-pro

Generate or edit images via the Gemini 3 Pro Image API.

Community
Intermediate
SJTU-IPADSSJTU-IPADS
7

image-utils

Automate deterministic image editing tasks on image collections with Pillow.

Official
Intermediate
pmcdowallpmcdowall

clip

Generate shared embeddings for cross-modal image and text retrieval.

Community
Intermediate
jykimjykim
49

image-generation-skill

Generate 1024x1024 images via DALL-E 3 for documents and slide decks.

Community
Advanced