Fetching the paper…

MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents · Around