Perfect score on IPhO 2025 theory by Gemini agent
Organizations: Santa Clarita, California 91350, USA
Abstract
The International Physics Olympiad (IPhO) is the world's most prestigious and renowned physics competition for pre-university students. IPhO problems require complex reasoning based on deep understanding of physical principles in a standard general physics curriculum. On IPhO 2025 theory problems, while gold medal performance by AI models was reported previously, it falls behind the best human contestant. Here we build a simple agent with Gemini 3.1 Pro Preview. We run it five times and it achieved a perfect score every time. However, data contamination could occur because Gemini 3.1 Pro Preview was released after the competition.
Figures & tables
| event | accuracies |
|---|---|
| International Mathematical Olympiad (IMO) | 83.3% [ 5 , 6 , 7 , 8 ] |
| International Physics Olympiad (IPhO) | 78.3% [ 9 ] , 75.7% [ 10 ] , 71.7% [ 11 ] , 77.3% [ 12 ] , |
| 72.7%, 84% [ 13 ] , 87.7% [ 14 ] , 100% [this work] | |
| International Chemistry Olympiad (IChO) | 75.4% [ 15 ] , 82.8% [ 14 ] |
| International Olympiad in Informatics (IOI) | 74.4% [ 16 ] , 88.9% [ 17 ] |
| International Olympiad on | 85.6% (theory), 88.5% (data analysis) |
AI Agent Evaluation
AI agent evaluation measures whether tool-using systems complete tasks reliably, safely, and efficiently across interactive settings.
See all 519 papers