Benchmark explainer
What is MATH level 5?
Competition math; L5=hardest band
MATH Level 5 — the hardest (level-5) competition-mathematics problems from the MATH dataset; scored as accuracy (%).
How it's scored
- Metric
- Exact-match boxed answer (MATH-500 common)
- Score ceiling
- 100
- Construction
- 12.5k problems from competition archives, difficulty-tagged
- Human baseline
- IMO gold ~90%; CS PhD non-specialist ~40%
How trustworthy
Discrimination
100/100Does it still separate models?
Saturation headroom
89.6/100How far from ceiling / clustered at the top?
Contamination resistance
0/100Public vs held-out; training-leak risk.
Harness comparability
100/100Is it apples-to-apples, or mixed / vendor-optimized?
Freshness
7.5/100How old is the benchmark?
Contamination history: Public since 2021, heavily contaminated/memorized
Limitations: Rewards guessed answers; all public; code interpreter trivializes many
Who leads MATH level 5
| Model | Score | Evidence |
|---|---|---|
| GPT-5 | 98.13 | reproduced |
| GPT-5 mini | 97.85 | reproduced |
| o4-mini | 97.83 | reproduced |
| o3 | 97.77 | reproduced |
| Claude Sonnet 4.5 | 97.73 | reproduced |
Compare the top MATH level 5 scorers
Frequently asked questions
MATH level 5 (MATH (competition mathematics)): MATH Level 5 — the hardest (level-5) competition-mathematics problems from the MATH dataset; scored as accuracy (%).
GPT-5 leads MATH level 5 at 98.13 — independently reproduced. The full leaderboard above lists every recorded measurement, not just the headline number.
85 models have recorded scores on MATH level 5, spanning a score spread of 94.84.
tensor.news grades MATH level 5 B for integrity (score 73/100), ranking #50 of 65 benchmarks we assess across discrimination, saturation, contamination resistance, harness comparability, and freshness.
Yes — top scores are clustering near the ceiling (saturation ratio 0.11), so MATH level 5 no longer separates leading models well. Treat small gaps at the top with caution.
Its test set is public, so contamination risk is on the table: a high score may partly reflect training-data overlap rather than capability.
Scores are reported under a consistent harness, so comparisons on MATH level 5 are reasonably apples-to-apples.
Every MATH level 5 measurement is source-backed and tagged reproduced, self-reported, or unverified; the underlying record cites arxiv.org/abs/2103.03874.