Frontiercode
The Frontier
Best score over time · one chart, every benchmark
Full rankings
32 models tested · sorted by score
| # | Model | Score |
|---|---|---|
| 1 | 54.6 | |
| 2 | 53.5 | |
| 3 | 53.4 | |
| 4 | 53.3 | |
| 5 | 52.1 | |
| 6 | 50.9 | |
| 7 | 48.0 | |
| 8 | 47.5 | |
| 9 | 46.5 | |
| 10 | 44.2 | |
| 11 | 43.6 | |
| 12 | 42.7 | |
| 13 | 42.4 | |
| 14 | 41.3 | |
| 15 | 41.2 | |
| 16 | 40.1 | |
| 17 | 39.8 | |
| 18 | 38.5 | |
| 19 | 34.4 | |
| 20 | 31.8 | |
| 21 | 30.1 | |
| 22 | 28.6 | |
| 23 | 27.0 | |
| 24 | 26.6 | |
| 25 | 24.5 | |
| 26 | 24.3 | |
| 27 | 18.8 | |
| 28 | 17.6 | |
| 29 | 14.7 | |
| 30 | 13.6 | |
| 31 | 10.2 | |
| 32 | 8.0 |
Score distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Benchmarks that track with Frontiercode
Pearson correlation across models scored on both benchmarks. Closer to 1 = strongly predictive.
Frequently asked
About Frontiercode
What does Frontiercode measure?
Frontiercode is a knowledge benchmark in the BenchGecko catalog. 32 AI models have been tested on it. Scores range from 8.0 to 54.6 out of 100.
Which model leads on Frontiercode?
Claude Opus 5.5 from Anthropic leads Frontiercode with a score of 54.6. The median score across 32 tested models is 40.0.
Is Frontiercode saturated?
No · the top score is 54.6 out of 100 (55%). There is still meaningful room for improvement on Frontiercode.
Does Frontiercode predict performance on other benchmarks?
Yes · Frontiercode scores correlate 0.98 with Osworld 2 0 across 7 shared models. Models that do well on Frontiercode tend to do well on Osworld 2 0.
How often is Frontiercode data refreshed?
BenchGecko pulls updates daily. New model scores on Frontiercode appear as soon as they are published by Epoch AI or the model provider.
- Category
- Knowledge
- Max score
- 100
- Models
- 32
- Updated
- 2026-09-28
Top on Frontiercode
Claude Opus 5.5 · 54.6Claude Fable 5 · 53.5Claude Opus 5 · 53.4GPT-6 Astra · 53.3Claude Sonnet 5.5 · 52.1More knowledge benchmarks
Same category · related evaluations