Post by @cointelegraph • Hey

The scientists developed a tool called "AgentBench" to benchmark LLM models as agents. https://cointelegraph.com/news/chatgpt-and-claude-are-becoming-capa