Open platform for evaluating large language models (LLMs) called LMSYS Chatbot Arena. It discusses how the platform works and ranks various LLMs. Users can choose two models to chat with and then vote on which one provided a better response. LMSYS Chatbot Arena collects this data to compute an LLM Elo leaderboard. The article also details some of the LLMs that are available on the platform, including Claude by Anthropic, GPT-4-Turbo by OpenAI, and GPT-4o by OpenAI.
