Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-5 nano | o1-mini |
|---|---|---|
| GPQA Diamond | 71.2 | 60 |
| Humanity's Last Exam | 8.7 | — |
math
| Benchmark | GPT-5 nano | o1-mini |
|---|---|---|
| AIME 2025 | 85.2 | — |
| FrontierMath | 9.6 | — |
Category Winners
math
39.0
reasoning
34.2