https://github.com/jcmpagel/llmimage
Wikimedia + GEMINI AI
https://github.com/jcmpagel/llmimage
ai gemini wikimedia wikipedia
Last synced: 7 months ago
JSON representation
Wikimedia + GEMINI AI
- Host: GitHub
- URL: https://github.com/jcmpagel/llmimage
- Owner: jcmpagel
- Created: 2025-03-03T10:10:15.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2025-03-09T14:04:16.000Z (7 months ago)
- Last Synced: 2025-03-09T14:31:00.942Z (7 months ago)
- Topics: ai, gemini, wikimedia, wikipedia
- Language: JavaScript
- Homepage: https://llm.jonathanpagel.com/
- Size: 68.4 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Wikimedia Image-Enhanced Q&A Tool
This web application uses Google's Gemini AI to create rich, visual answers to questions by automatically retrieving and incorporating relevant images from Wikimedia Commons.
## Features
- **AI-powered Image Search**: Automatically generates search terms from your question to find relevant images
- **Visual Responses**: Creates comprehensive answers with embedded, relevant images
- **Shareable Results**: Create permanent links to share your Q&A results with others
## How It Works
1. **Search Term Generation**: When you submit a question, Gemini AI intelligently generates relevant search terms
2. **Image Retrieval**: These search terms are used to find appropriate images from Wikimedia Commons
3. **Image Analysis**: Gemini analyzes the images alongside your question to create a visual response
4. **Response Generation**: The AI creates a comprehensive answer that naturally incorporates relevant images with explanatory text
## Getting Started
1. Visit the tool at [https://llm.jonathanpagel.com/](https://llm.jonathanpagel.com/)
2. Type your question in the text box
3. Optionally add your own Gemini API key (get one for free from [Google AI Studio](https://aistudio.google.com/apikey))
4. Choose your preferred Gemini model:
- **Gemini 2.0 Flash**: Faster responses (default)
- **Gemini 2.0 Flash-Thinking**: More thoughtful responses with better reasoning
5. Click "Ask Question" and wait for your visual answer!
## Example Questions
The tool works best with questions that can benefit from visual aids:
- "Show me different types of cloud formations"
- "Explain the perceptron in machine learning"
- "How does the human heart work?"
- "What are the main architectural styles throughout history?"
- "Describe the life cycle of a butterfly"
## Technical Details
- **Frontend**: Pure HTML, CSS, and JavaScript
- **AI**: Google Gemini 2.0 models for text and image analysis
- **Image Source**: Wikimedia Commons API
- **Data Storage**: Supabase for storing and sharing responses