Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Hermes 3 70B | o3 |
|---|---|---|
| GPQA Diamond | 66.1 | 83.3 |
| Humanity's Last Exam | — | 14.7 |
| ARC-AGI v2 | — | 6.5 |
math
| Benchmark | Hermes 3 70B | o3 |
|---|---|---|
| AIME 2025 | — | 86.4 |
| FrontierMath | — | 15.8 |
coding
| Benchmark | Hermes 3 70B | o3 |
|---|---|---|
| SWE-bench Verified | — | 69.1 |
multimodal
| Benchmark | Hermes 3 70B | o3 |
|---|---|---|
| MMMU | — | 82.9 |
| MMMU-Pro | — | 76.4 |
| CharXiv Reasoning | — | 78.6 |
agent
| Benchmark | Hermes 3 70B | o3 |
|---|---|---|
| BrowseComp | — | 49.7 |
Category Winners
coding
58.9
math
43.4
reasoning
N
Hermes 3 70B46.9
multimodal
72.9
agent
22.9