Frontierswe
The Frontier
Best score over time · one chart, every benchmark
Full rankings
13 models tested · sorted by score
| # | Model | Score |
|---|---|---|
| 1 | 65.5 | |
| 2 | 62.3 | |
| 3 | 61.9 | |
| 4 | 56.3 | |
| 5 | 52.0 | |
| 6 | 47.0 | |
| 7 | 32.2 | |
| 8 | 30.2 | |
| 9 | 25.9 | |
| 10 | 25.3 | |
| 11 | 20.3 | |
| 12 | 19.6 | |
| 13 | 15.8 |
Score distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Benchmarks that track with Frontierswe
Pearson correlation across models scored on both benchmarks. Closer to 1 = strongly predictive.
Frequently asked
About Frontierswe
What does Frontierswe measure?
Frontierswe is a knowledge benchmark in the BenchGecko catalog. 13 AI models have been tested on it. Scores range from 15.8 to 65.5 out of 100.
Which model leads on Frontierswe?
GPT-6 Astra from OpenAI leads Frontierswe with a score of 65.5. The median score across 13 tested models is 32.2.
Is Frontierswe saturated?
No · the top score is 65.5 out of 100 (66%). There is still meaningful room for improvement on Frontierswe.
Does Frontierswe predict performance on other benchmarks?
Yes · Frontierswe scores correlate 0.97 with Artificial Analysis · CritPt across 8 shared models. Models that do well on Frontierswe tend to do well on Artificial Analysis · CritPt.
How often is Frontierswe data refreshed?
BenchGecko pulls updates daily. New model scores on Frontierswe appear as soon as they are published by Epoch AI or the model provider.
- Category
- Knowledge
- Max score
- 100
- Models
- 13
- Updated
- 2026-09-28
Top on Frontierswe
GPT-6 Astra · 65.5Claude Opus 5.5 · 62.3Claude Sonnet 5.5 · 61.9Claude Fable 5.1 · 56.3Claude Opus 5 · 52.0More knowledge benchmarks
Same category · related evaluations