Primary evidence
- Advanced version of Gemini with Deep Think officially achieves gold-medal standard at the International Mathematical OlympiadS2 · Official institutional · Official announcement · 2025-07-21
Gemini Deep Think reaches officially graded IMO 2025 gold-medal standard in natural language
Unlike AlphaProof and AlphaGeometry 2 in 2024, the 2025 system consumed official problem statements directly and produced natural-language proofs without manual translation into Lean or another domain-specific formal language. It completed the task within the 4.5-hour competition limit.
What happened?
An advanced Gemini Deep Think model solved five of the six IMO 2025 problems for 35 out of 42 points, reaching gold-medal standard; Google DeepMind reported that IMO graders evaluated the natural-language solutions under the same criteria used for contestants.
Why it matters
This was a major change in interface and latency, showing that a general reasoning model could reach elite Olympiad performance end-to-end in the language used by human contestants.
Significance rationale: The result moved Olympiad-level AI from the manually formalized, multi-day AlphaProof workflow of 2024 to end-to-end natural-language proof generation within the contest time limit, with grading certified by IMO coordinators.
Historical context
Evidence & status at a glance
- Significance
- H1 · Historical milestone
- Verification
- Independently corroborated
- Evidence level
- E2 · Research record
- AI role
- AI primary, human verified
- Formal assurance
- No formal assurance
Contribution & verification
- Mathematical novelty
- Benchmark result
- Interface
- Informal
- AI contribution
- Gemini generated the solutions end-to-end from the natural-language problems; external IMO graders assessed the submitted solutions.
- Human contribution
- DeepMind researchers developed and post-trained the model and supplied general problem-solving guidance; IMO coordinators and graders provided the competition problems and external scoring.
- Verification
- Independently corroborated · E2 · Research record · No formal assurance
- Last reviewed
- 2026-09-11
Technical details
- Systems
- gemini-deep-think
- Organizations
- google-deepmind, international-mathematical-olympiad
- Problems
- international-mathematical-olympiad-2025
- Methods
- reasoning-model, parallel-thinking, reinforcement-learning, natural-language-proof
Sources & artifacts
- Advanced version of Gemini with Deep Think officially achieves gold-medal standard at the International Mathematical OlympiadS2 · Official institutional · Official announcement · accessed 2026-09-10
- OpenAI and Google outdo the mathletes, but not each otherS4 · Reputable secondary · Media report · accessed 2026-09-10
Verification history
- 2025-07-21 · Partially verified · E2 · Research record
Google DeepMind published the score and quoted the IMO president confirming 35/42 and gold-medal standard after official grading.
- 2026-09-10 · Independently corroborated · E2 · Research record
The Chronicle records the result as independently corroborated at the event level: IMO coordinators graded the submission, and independent reporting confirmed the 35/42 gold-standard result and the distinction between Google's officially coordinated evaluation and OpenAI's separate evaluation. This status concerns the competition event, not a broader claim about mathematical correctness beyond the grading process.
Corrections
No corrections recorded.