SuperAPI AI Directory
Partner with us中文AI newsFavorites
All toolsAI imagesNano-Bananary

Nano-Bananary

Nano-Bananary (Banana Supermarket) is an AI image and video generation tool based on Gemini 2.5 Flash Image, open-sourced on GitHub. It focuses on "no prompt needed" one-click playstyle generation, with a variety of built-in presets, allowing users to quickly generate images without writing prompts. The tool supports local brushing and selection, enabling editing of specific areas of an image; it supports continuous editing, where each generated output can be directly used as input for the next edit or generation, achieving iterative creation. It also provides a history feature, making it easy to call previously generated images as new input. The interface supports one-click switching between Chinese and light themes, and a video generation feature has been launched, allowing each playstyle to be converted into a video with one click. It also includes comparison viewing features: enlarged preview, horizontal comparison and comparison image download, slider comparison, and a custom prompt module. The project requires a Node.js environment, installs dependencies via npm, configures GEMINI_API_KEY in .env.local, and runs npm run dev to start. It can be used online in AI Studio.

Open website

About this tool

Nano-Bananary (Chinese name "Banana Supermarket") is an AI creative generation project open-sourced on GitHub by developer ZHO-ZHO-ZHO, based on the Gemini 2.5 Flash Image model, positioned as "one-click generation for various playstyles, no prompt needed, supports local brushing and selection, continuous editing".

Purpose

The project is aimed at users who want to quickly experience a variety of AI image/video playstyles without writing prompts. Through preset playstyles, users can directly generate images or videos, lowering the barrier to use. The project is also suitable for creative scenarios that require continuous iterative editing and local image modification.

Main capabilities

  • One-click generation: built-in multiple playstyles, one-click generation without prompts.
  • Local brushing and selection: supports brushing and selecting local areas of an image for editing.
  • Continuous editing: each output can be directly used as input for the next edit or generation, enabling chain creation.
  • History: added a history feature, making it convenient to directly use previously generated images as input.
  • Video generation: all playstyles can be converted into video with one click (added on 2025-09-16).
  • Interface and theme: one-click switching between Chinese interface and light theme (added on 2025-09-16).
  • Comparison features: includes enlarged preview, horizontal comparison and direct download of comparison images, slider comparison (updated on 2025-09-05).
  • Custom prompts: provides a custom prompt module for personalized generation (added on 2025-09-05).
  • Playstyle library update: the main playstyles of the "🍌 Prompt Library" have been updated (2025-09-06).

Usage

This project is a locally run application. The prerequisite is Node.js. Install dependencies: npm install. Configure GEMINI_API_KEY in .env.local as your Gemini API key. Run the application: npm run dev. It can also be used online via AI Studio, link: https://ai.studio/apps/drive/1JknFrFFdiOm7FIA8MLOJa_vtJN2g24c1。

Applicable scenarios

Suitable for users who need to quickly generate images/videos and do not want to manually write prompts; suitable for creators who need local modifications and continuous iterative editing; suitable for scenarios where you want to compare the original image with the generated result and directly share the comparison image.

Limitations

The project depends on a Gemini API key, which must be configured by yourself. Running requires a Node.js environment. Other information such as specific charges, pricing, and platform support scope is not currently provided.

Related tools

Browse category