Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Nemotron Labs Dory 1.0 (4B) | o3-mini |
|---|---|---|
| GPQA Diamond | 72.4 | 77.2 |
| Humanity's Last Exam | 10.9 | — |
coding
| Benchmark | Nemotron Labs Dory 1.0 (4B) | o3-mini |
|---|---|---|
| SWE-bench Verified | 59.5 | 49.3 |
| SciCode | 34.9 | — |
general
| Benchmark | Nemotron Labs Dory 1.0 (4B) | o3-mini |
|---|---|---|
| SimpleQA | — | 15 |
agent
| Benchmark | Nemotron Labs Dory 1.0 (4B) | o3-mini |
|---|---|---|
| TAU-Bench Retail | — | 57.6 |
math
| Benchmark | Nemotron Labs Dory 1.0 (4B) | o3-mini |
|---|---|---|
| FrontierMath | — | 9.2 |
Category Winners
coding
27.2
math
16.1
reasoning
67.7
general
18.5
agent
21.9