The LMSYS Chatbot Arena serves as the primary benchmark for Large Language Models (LLMs), using a crowdsourced, blind pairwise comparison methodology to calculate Elo ratings. While the platform has evolved to support multimodal evaluation—including vision-language
