01AI-Powered Search with access to multiple AI models (GPT-5.1, LLaMA 3.1, Claude, Gemini)
020 GitHub stars
03Automatic Request Queuing and requeuing when rate limits are hit, with priority-based retry
04Robust Error Handling with exponential backoff and up to 10 retry attempts
05Smart Rate Limiting with a global 5 RPM limit, token bucket algorithm, and automatic retries
06MCP Compatible, ensuring seamless integration with clients like Claude Desktop and Cursor