Rank LLMs, RAG systems, and prompts using automated head-to-head evaluation - View it on GitHub