Perfect score on IPhO 2025 theory by Gemini agent

TL;DR

This paper presents a simple AI agent based on Gemini 3.1 Pro Preview that achieved perfect scores on IPhO 2025 theory problems, demonstrating significant progress in physics problem-solving capabilities.

Contribution

The authors develop and test a new AI agent that attains perfect scores on IPhO 2025 problems, surpassing previous AI performance and approaching human-level reasoning.

Findings

AI agent achieved perfect scores five times in testing.

Performance suggests potential for advanced physics reasoning by AI.

Data contamination risk due to model release timing.

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.