Gemini Gradio Image Editor & Analyzer

A Gradio web interface that uses the Google Gemini API to perform powerful AI-based image editing and analysis. This tool can transform your photos into different art styles, add or remove objects, or simply provide a detailed text description of an image.

Example: Make him hold a giant trophy. OR Describe this scene

Features

• AI Image Editing: Provide an image and a text prompt (e.g., "Make this a watercolor painting," "add a superhero cape to the person") to generate a new version of your image.
• AI Image Analysis: Ask questions about an image (e.g., "describe this scene in detail," "what kind of event is this?") to receive a textual analysis from the AI.
• Dual-Mode Interface: The UI intelligently displays either the generated image (with a download button) or the generated text, depending on the model's response.
• Simple & Clean UI: Built with Gradio for a straightforward and responsive user experience.

Setup

1. Clone the repository:

   git clone https://github.com/zakcali/gemini-gradio-image-editor.git
   cd gemini-gradio-image-editor

2. Install the required packages. It is recommended to use a virtual environment.

   pip install -r requirements.txt

   Your requirements.txt file should contain:

   gradio
   google-genai
   pillow
   

Configuration

Before running the application, you must set your Google Gemini API key as an environment variable. You can get a key from Google AI Studio.

# On macOS or Linux
export GEMINI_API_KEY="YOUR_API_KEY_HERE"

# On Windows
set GEMINI_API_KEY="YOUR_API_KEY_HERE"

Running the Application

Run the main script from your terminal:

python gradio-edit.py

The application can be launched in several ways:

To run locally on your machine:

The script defaults to this. Gradio will provide a local URL like http://127.0.0.1:7860.

demo.launch()

To serve to other devices on your local network (LAN):

Modify the last line in the script to include your machine's local IP address.

# Replace "192.168.1.XX" with your actual LAN IP address
demo.launch(server_name="192.168.1.XX", server_port=7860)

To share publicly over the internet:

Set the share parameter to True. Gradio will generate a temporary public URL for you.

demo.launch(share=True)

For more details, see the official Gradio guide on Sharing Your App.

How It Works

This application sends the user-uploaded image and text prompt to the Gemini API, specifically using the gemini-2.5-flash-image-preview model. As of 6 Sep 2025, this is Nano Banana model. The Gemini model is multimodal and can understand both the image and the text context.

Based on the prompt, it will either:

1. Return a new image as inline_data, which the script then saves as a PNG file and displays.
2. Return a text response, which the script displays in a separate text box.

The script gracefully handles both response types, making it a versatile tool for both creative editing and analytical tasks.