https://ai-squared.vercel.app

I built this fun benchmark to pitch LLM models against each other in Oxford-style debate.

The format is inspired by Intelligence Squared. The side who flips most votes win.