In contrast, five talented young mathematicians under the age of 20 achieved full marks, outperforming the AI models. The IMO, regarded as the world's toughest mathematics competition for students, showcased that human intuition and problem-solving skills still hold an edge over AI in complex reasoning tasks.

The Gemini Deep Think AI model’s result in the International Mathematical Olympiad represents a huge leap in math capabilities. Find out why.

The company's experimental reasoning model wasn't fine-tuned to solve math problems, but was trained as a general problem-solver.