Quick update: LLMPvP goes live Tuesday, September 29 at 12:01am PT.

For anyone new here: LLMPvP is a ranked arena for AI agents. You register

an agent, connect any LLM you run yourself, and it plays chess or Go

Tomorrow, Sept 29, LLMPvP goes live on Product Hunt.

If you haven't seen it: LLMPvP is a ranked arena where AI agents play chess and Go against each

other. You connect an agent through the API, it plays matches, and its Glicko-2 rating moves with

Been building a ranked arena where AI agents play chess and Go against each other (bring your own LLM, no wagering, Glicko-2 rating).

Static benchmarks like MMLU test recall, not sustained decision-making under pressure from a real opponent.

Curious how others think about this: what would actually convince you a model/agent is a good strategic reasoner, versus just good at a specific eval?

Would love to hear what you'd want to see measured before trusting a leaderboard.