International Physics Olympiad 2025 (Theory) (IPhO 2025 (Theory))
We show this table for reference; we do not rank on it.
The three official theory problems from the 2025 International Physics Olympiad, scored with blinded human evaluation.
Benchmark score on IPhO 2025 (Theory) — November 13, 2026
We compile the IPhO 2025 (Theory) rows from secondary reports. GPT-5.4 Pro leads the table at 93.5%. We do not use these results to rank models overall.
1 modelMathematicsCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (1 model)
The one result on this page has no documented run by the benchmark or an independent evaluator: 1 reported in another publisher’s document.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | GPT-5.4 ProOpenAIReported by Meta AI (comparison table) | 93.5% | Reported by Meta AI (comparison table) | Not reported | Closed |
- Lab-reported
- Benchmark-run
- Independent
About IPhO 2025 (Theory)
Year
2026
Tasks
3 olympiad theory problems
Format
Physics olympiad theory
Difficulty
International olympiad physics
Meta reports IPhO 2025 Theory as a Contemplating-mode comparison benchmark, scored with human evaluation guided by the official olympiad rubric. BenchLM stores it as a display-only advanced physics benchmark.
Freshness and provenance
Version
IPhO 2025 (Theory) 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.
Questions
What does IPhO 2025 (Theory) measure?
The three official theory problems from the 2025 International Physics Olympiad, scored with blinded human evaluation.
Which model scores highest on IPhO 2025 (Theory)?
GPT-5.4 Pro by OpenAI currently leads with a score of 93.5% on IPhO 2025 (Theory).
How many models are evaluated on IPhO 2025 (Theory)?
1 AI models have published results on IPhO 2025 (Theory) in the BenchLM catalog.
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.