Greg Kamradt Responds to Evaluation Dispute: Same Rolling Window Used for Opus and OpenAI

GregKamradt · x · 2026-07-30

Addressing the recent controversy over Claude Opus allegedly cheating in the ARC-AGI evaluation, Greg Kamradt clarified the methodology. He stated that they used the exact same "rolling window" convention for both Anthropic and OpenAI models to ensure fairness in the test.

Related event: ARC-AGI Open Sources Eval Code Amid Testing Dispute(2 posts)→

Original post →

More from Models

Models channel →