This project provides scripts to benchmark Large Language Models (LLMs) from two sources: local models via LM Studio and API-based models via OpenRouter. It measures performance metrics like load time, generation speed, and resource consumption (power/memory).