ZeroLu/awesome-nanobanana-pro

What it solves

This repository provides a curated library of high-fidelity prompts designed to help users get the best possible results from the Nano Banana AI model. It addresses the challenge of crafting complex, specific inputs needed to achieve professional-grade visual effects, such as photorealistic portraits, era-specific aesthetics, and consistent character rendering.

How it works

The project acts as a community-driven collection of "recipes" for AI image generation. Users can browse categorized prompts—ranging from hyper-realistic crowd scenes to 2000s-style mirror selfies—and copy the exact text or JSON structures provided. These prompts include specific instructions on lighting, camera settings, lens types, and subject details to guide the model toward high-quality, consistent outputs.

Who it’s for

This repository is for AI enthusiasts, prompt engineers, and creative professionals who use the Nano Banana model for visual experiments. It is particularly useful for those looking to improve their image consistency, master specific artistic styles, or generate professional-looking content like business headshots or cinematic photography.

Highlights

• Diverse Categories: Includes prompts for photorealism, interior design, e-commerce, social media, and photo restoration.
• Technical Precision: Prompts often specify camera hardware (e.g., Sony A7III, 85mm lens) and lighting setups (e.g., three-point lighting) to ensure professional results.
• Structured Formats: Provides both descriptive text prompts and structured JSON objects for more complex, multi-layered image generation tasks.
• Community Sourced: Features high-quality inputs gathered from platforms like X (Twitter), WeChat, and Replicate to ensure the most effective techniques are shared.