Anthropic only said that the model performed 35,000 actions to reach the last gym leader, Surge. Last week, a researcher tried out an early preview of Claude 3.7 Sonnet. The results were striking.

AI Benchmarks Under Fire: 'Pokémon' Games Expose Cracks in Model Amir Balam/Unsplash While the viral claim stirred excitement, it conveniently left out a crucial detail: Gemini had a leg up.