Chinese AI Models Achieve Perfect Scores at World's Toughest Mathematics Competition

Artificial intelligence has reached another major milestone, with two Chinese technology companies announcing that their AI models achieved perfect scores at the 2026 International Mathematical Olympiad (IMO)—one of the world's most prestigious and demanding mathematics competitions. For the first time, AI systems have earned full marks under the IMO's official evaluation process, matching the performance of the competition's highest-scoring human participants.

The achievement was announced by Huawei and Xiaohongshu, marking a significant advancement in AI's mathematical reasoning capabilities.

A Historic First for Artificial Intelligence

The companies revealed that their respective AI models—Huawei's Celia and Xiaohongshu's dots-note-3.0—each scored 42 out of 42, successfully solving every problem presented in this year's Olympiad. According to Xiaohongshu, no large language model had previously achieved a perfect score through the IMO's official judging process. The company described the accomplishment as an important breakthrough in AI's ability to perform complex mathematical reasoning without human assistance.

Building on Rapid AI Progress

The latest achievement builds on the rapid progress AI has made in competitive mathematics over the past two years. In 2024, Google's AI model achieved a silver medal-level performance after solving four of the six IMO problems over several days. By 2025, AI models developed by Google and OpenAI reached gold medal-level scores. 

However, they still fell short of the perfect scores achieved by the competition's top human contestants. The 2026 results suggest that AI systems have now closed that gap.

Strict Competition Conditions

This year's International Mathematical Olympiad was held in Shanghai, bringing together 666 students from around the world. Only seven contestants achieved perfect scores, according to the official competition results.

To ensure fairness, participating AI companies received the examination questions only after the human competition had concluded. Their models were required to submit solutions within a fixed time limit, with no human intervention permitted during testing. The completed solutions were then evaluated using the same official grading standards applied to student participants.

Experts Say AI Has Crossed Another Milestone

Huawei stated that its AI model Celia demonstrated strong reasoning abilities across multiple branches of mathematics, highlighting its capacity to solve complex Olympiad-level problems.

Meanwhile, Xiaohongshu noted that dots-note-3.0 was participating in the IMO evaluation for the first time. Adding to the momentum, Deedy Das, a partner at Menlo Ventures, independently tested four leading AI models—including systems from OpenAI, Anthropic, Axiom Math, and China's Moonshot AI—using the same IMO questions. According to Das, each model also achieved a perfect score of 42 out of 42.

Describing the breakthrough on LinkedIn, he wrote that "the frontier of AI has officially moved well past IMO math."

Why This Matters

For decades, the International Mathematical Olympiad has represented one of the highest benchmarks of human mathematical reasoning. AI's ability to now achieve perfect scores under official evaluation conditions signals a new phase in artificial intelligence—one where advanced reasoning, rather than simple information retrieval, is becoming increasingly sophisticated. While these systems are not replacing human creativity, their growing ability to solve highly abstract problems is likely to reshape research, education, engineering, and scientific discovery.

Editor's Note

Mathematics has long been viewed as one of the clearest expressions of human reasoning—a discipline where creativity, logic, and intuition converge. AI's performance at this year's International Mathematical Olympiad does not diminish that achievement; instead, it redefines the conversation.

The question is no longer whether AI can solve difficult problems, but how society chooses to use these increasingly capable systems. As artificial intelligence moves from assisting calculations to demonstrating advanced reasoning, education may need to shift its focus from producing correct answers to cultivating curiosity, critical thinking, ethical judgment, and original ideas—qualities that remain deeply human.

Frequently Asked Questions (FAQs)

1. What is the International Mathematical Olympiad (IMO)?

The IMO is the world's most prestigious international mathematics competition for students under the age of 20, featuring some of the most challenging mathematical problems.

2. Which AI models achieved perfect scores?

Huawei's Celia and Xiaohongshu's dots-note-3.0 both achieved perfect scores under the IMO's official judging process.

3. Why is this achievement significant?

It marks the first time AI models have earned full marks in the International Mathematical Olympiad using the competition's official evaluation standards.

4. How many human participants scored full marks in 2026?

Out of 666 contestants, only seven students achieved perfect scores.

5. Were humans involved in solving the AI's answers?

No. According to the participating companies, testing was conducted without any human intervention, and the models submitted their own solutions within the official time limit.

6. How has AI's performance improved over recent years?

In 2024, Google's AI achieved a silver medal-level score. In 2025, Google and OpenAI achieved gold medal-level performance. In 2026, multiple AI models achieved perfect scores.

7. Which other AI models reportedly scored full marks?

Investor Deedy Das reported that models from OpenAI, Anthropic, Axiom Math, and Moonshot AI also achieved 42 out of 42 when independently tested using the same questions.

8. What does this mean for education and research?

The milestone suggests AI is becoming increasingly capable of advanced reasoning, which could transform scientific research, engineering, education, and mathematical problem-solving while prompting educators to place greater emphasis on creativity and critical thinking.