A multi-model interface where LLMs debate with each other
Orchestrates real-time skepticism between models to catch hallucinations before you see them.

Debate format tests persuasion under opposition, not just completion quality like LMSys Arena.
AI researchers, ML engineers, LLM enthusiasts
LMSys Chatbot Arena · HELM Benchmark
The format is inspired by Intelligence Squared. The side who flips most votes win.
Orchestrates real-time skepticism between models to catch hallucinations before you see them.
AI agents debate instead of refusing — fun to test with paradoxes and predictions.
Multi-agent code review with internal debate beats single-pass LLM tools.
Debate mode where models change minds is novel, but model comparison tools already exist.
Yet another AI directory, but the free SEO tools are actually useful.
Side-swapped debate matchups expose model weaknesses standard benchmarks miss.