MCP Playwright Multi-LLM Benchmark

A modular AI benchmarking framework built with TypeScript + Playwright.

This project compares multiple LLM providers using a unified interface and measures their responses to the same prompt.

🎯 Purpose

Ask the same question to multiple AI providers:

"What is Playwright?"

And compare:

• Response output
• Execution time (ms)
• Architecture consistency
• Extensibility for future evaluation metrics

🚀 Supported Providers

• OpenAI
• Ollama
• Gemini
• Claude

All providers implement a common interface

🚀 Run the Benchmark

npx playwright test tests/benchmark/llm-benchmark.spec.ts

🚀 Prerequisites

•	Node.js v18+
•	npm
•	(Optional) API keys when real providers are connected