LLM Comparator Extension

A Chrome extension that allows you to send prompts to multiple AI models (ChatGPT, Claude, AskMe) simultaneously and get AI-powered summaries with two analysis options: local browser LLM or OpenAI API.

🎯 Features

• Multi-Model Comparison: Send the same prompt to ChatGPT, Claude, and AskMe simultaneously
• Dual Analysis Options:
  • Local Browser LLM: Intelligent analysis engine running entirely in your browser (free, private)
  • OpenAI API: Cloud-based analysis using GPT-3.5-turbo (higher quality, requires API key)
• Intelligent Comparative Analysis: Get structured insights with similarities, differences, and consolidated answers
• Complete Privacy Option: Local analysis never sends data externally
• Smart UI Controls: Settings adapt based on your chosen analysis method
• One-Click Tab Management: Automatically open AI service tabs

🚀 Setup

1. Load Extension: Load the extension in Chrome (Developer mode)
2. Open AI Service Tabs: Use the "Open Tab" buttons or manually navigate to:
   • ChatGPT: https://chatgpt.com
   • Claude: https://claude.ai
   • AskMe: https://askme.mobileye.com
3. Choose Analysis Method:
   • For Local Analysis: No setup required - works immediately offline
   • For OpenAI Analysis: Configure your OpenAI API key in the extension popup

📖 Usage

Basic Workflow

1. Select Models: Check the boxes for the AI models you want to query
2. Enter Prompt: Type your question/prompt in the text area
3. Send: Click "Send" - the extension will automatically:
   • Send your prompt to all selected models
   • Collect their responses in real-time
   • Show analysis options when responses are ready

Analysis Options

🏠 Local Browser LLM (Recommended)

• Setup: None required
• Cost: Completely free
• Privacy: 100% local - no data leaves your browser
• Features: Intelligent analysis with keyword extraction, pattern recognition, and quality assessment

☁️ OpenAI API

• Setup: Requires OpenAI API key
• Cost: Pay-per-use (typically fractions of a cent per summary)
• Privacy: Data sent to OpenAI for processing
• Features: Advanced language model analysis

Summary Analysis Includes

• Key Similarities: What all models agreed on
• Key Differences: Where models diverged in their approaches
• Unique Insights: Special perspectives each model contributed
• Quality Assessment: Which response was most comprehensive/accurate
• Consolidated Answer: Best elements combined into one cohesive response

🔧 Technical Features

Local Browser LLM Engine

• Intelligent Response Parsing: Automatically identifies and extracts model responses
• Keyword Analysis: Extracts and analyzes important themes and concepts
• Pattern Recognition: Identifies questions, examples, and structured content
• Complexity Measurement: Analyzes sentence structure and lexical diversity
• Quality Scoring: Multi-factor assessment of response comprehensiveness
• Sentence Consolidation: Combines best elements while avoiding duplication

Security & Privacy

• CSP Compliant: No external script dependencies
• Local Storage: API keys stored securely in Chrome's storage
• No External Dependencies: Local analysis works completely offline
• Selective Privacy: Choose between local processing or cloud analysis

🛠️ Configuration

OpenAI API Setup (Optional)

1. Get an API key from OpenAI
2. In the extension popup, select "OpenAI API (Cloud-based)" from the Summary Method dropdown
3. Enter your API key in the OpenAI settings section
4. The extension will securely store your key for future use

Browser LLM (Default)

• Auto-Initialize: Loads automatically when extension opens
• No Configuration: Works out of the box
• Offline Capable: Functions without internet connection
• Instant Analysis: No external API delays

💰 Costs

• Local Browser LLM: Completely free forever
• OpenAI API: Uses GPT-3.5-turbo model (~$0.001-0.002 per summary)

🔐 Privacy & Security

Local Analysis Mode

• ✅ Zero external data transmission
• ✅ Complete privacy - analysis happens in your browser
• ✅ No API keys required
• ✅ Works offline

OpenAI Analysis Mode

• ⚠️ Data sent to OpenAI for processing
• ✅ API key stored locally in Chrome's secure storage
• ✅ Direct browser-to-OpenAI communication
• ✅ No third-party data collection

🎨 UI Improvements

• Adaptive Interface: Settings automatically show/hide based on selected analysis method
• Better Layout: Summary controls positioned logically below model outputs
• Real-time Status: Progress tracking for local LLM initialization
• One-click Tab Management: Built-in buttons to open AI service tabs

🚀 Performance

• Local Analysis: Instant results, no network delays
• Smart Caching: Models stay loaded for repeated use
• Efficient Parsing: Optimized for real-time response processing
• Background Processing: Non-blocking analysis operations

🔄 Version History

• v1.1: Added local browser LLM with intelligent analysis engine
• v1.0: Initial release with OpenAI-powered summaries