FrontierMath-Tier-4-v2-Private
The Frontier
Best score over time · one chart, every benchmark
Full rankings
48 models tested · sorted by score
| # | Model | Score |
|---|---|---|
| 1 | 97.6 | |
| 2 | 95.0 | |
| 3 | 90.2 | |
| 4 | 87.8 | |
| 5 | 82.9 | |
| 6 | 80.5 | |
| 7 | 73.2 | |
| 8 | 70.7 | |
| 9 | 61.0 | |
| 10 | 58.5 | |
| 11 | 56.1 | |
| 12 | 49.0 | |
| 13 | 46.3 | |
| 14 | 46.0 | |
| 15 | 39.0 | |
| 16 | 36.6 | |
| 17 | 34.1 | |
| 18 | 31.7 | |
| 19 | 31.7 | |
| 20 | 31.7 | |
| 21 | 29.3 | |
| 22 | 29.3 | |
| 23 | 29.3 | |
| 24 | 26.8 | |
| 25 | 26.8 | |
| 26 | 26.8 | |
| 27 | 26.8 | |
| 28 | 25.6 | |
| 29 | 24.4 | |
| 30 | 24.4 | |
| 31 | 21.9 | |
| 32 | 21.9 | |
| 33 | 21.9 | |
| 34 | 19.5 | |
| 35 | 17.1 | |
| 36 | 17.1 | |
| 37 | 17.1 | |
| 38 | 14.6 | |
| 39 | 12.2 | |
| 40 | 12.2 | |
| 41 | 12.2 | |
| 42 | 9.8 | |
| 43 | 4.9 | |
| 44 | 4.9 | |
| 45 | 2.4 | |
| 46 | 2.4 | |
| 47 | 2.4 | |
| 48 | 2.4 |
Score distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Benchmarks that track with FrontierMath-Tier-4-v2-Private
Pearson correlation across models scored on both benchmarks. Closer to 1 = strongly predictive.
Frequently asked
About FrontierMath-Tier-4-v2-Private
What does FrontierMath-Tier-4-v2-Private measure?
FrontierMath-Tier-4-v2-Private is a knowledge benchmark in the BenchGecko catalog. 48 AI models have been tested on it. Scores range from 2.4 to 97.6 out of 100.
Which model leads on FrontierMath-Tier-4-v2-Private?
GPT-6 Astra from OpenAI leads FrontierMath-Tier-4-v2-Private with a score of 97.6. The median score across 48 tested models is 26.8.
Is FrontierMath-Tier-4-v2-Private saturated?
Yes · the top model on FrontierMath-Tier-4-v2-Private has reached 97.6 out of 100, within 5% of the theoretical ceiling. This benchmark is approaching saturation and may be replaced by a harder successor.
Does FrontierMath-Tier-4-v2-Private predict performance on other benchmarks?
Yes · FrontierMath-Tier-4-v2-Private scores correlate 0.96 with FrontierMath-Tier-4-2025-07-01-Private across 21 shared models. Models that do well on FrontierMath-Tier-4-v2-Private tend to do well on FrontierMath-Tier-4-2025-07-01-Private.
How often is FrontierMath-Tier-4-v2-Private data refreshed?
BenchGecko pulls updates daily. New model scores on FrontierMath-Tier-4-v2-Private appear as soon as they are published by Epoch AI or the model provider.
- Category
- Knowledge
- Max score
- 100
- Models
- 48
- Updated
- 2026-09-28
Top on FrontierMath-Tier-4-v2-Private
GPT-6 Astra · 97.6Claude Opus 5.5 · 95.0Claude Fable 5 · 90.2Claude Fable 5.1 · 87.8GPT-5.6 Sol · 82.9More knowledge benchmarks
Same category · related evaluations