Event / publication / experimentBranchesJuly 2025

AI reaching gold-medal level at the International Mathematical Olympiad

Google DeepMindOpenAI

This page delivers no verdict; it presents claims and evidence with their sources. No theory in science is absolute. Method

For everyone

In July 2025 Google DeepMind's Gemini Deep Think model solved five of the six International Mathematical Olympiad problems with natural-language proofs and was graded by official coordinators at gold-medal level (35 of 42). OpenAI reported the same score in its own evaluation. The models did not solve the sixth problem, regarded as the hardest.

For specialists

A year earlier (2024) AlphaProof and AlphaGeometry 2 reached silver level with the problems hand-translated into Lean. The 2025 results were obtained without tools, directly in natural language and within contest time. DeepMind's solutions were officially graded by IMO coordinators; OpenAI's score comes from its own panel of former medallists.

Claims

Evidence

Supporting

Measurement

IMO coordinators graded Gemini Deep Think's five solutions with the same criteria as student solutions and awarded full marks.[1]

Challenging

—

Open questions

Connections

Sources

  1. Web (2025). Advanced version of Gemini with Deep Think officially achieves gold-medal standard at the International Mathematical Olympiad. Google DeepMinddeepmind.googlearchive

Search the literature

Look for scholarly work on this topic in other indexes:

Cite this page

APA

Athena contributors. (2026). AI reaching gold-medal level at the International Mathematical Olympiad. Athena: The tree of human knowledge. Retrieved November 1, 2026, from https://latifkedi.github.io/athena/en/d/yapay-zeka-imo/

BibTeX
@misc{athena_yapay_zeka_imo,
  title        = {AI reaching gold-medal level at the International Mathematical Olympiad},
  author       = {{Athena contributors}},
  year         = {2026},
  howpublished = {\url{https://latifkedi.github.io/athena/en/d/yapay-zeka-imo/}},
  note         = {Athena: The tree of human knowledge. Accessed: 2026-11-01}
}

Added: September 27, 2026 · Last changed: September 27, 2026 · Page history

✎ Suggest a correction