Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-4o | Nemotron 3.5 Lightning (30B A3B) |
|---|---|---|
| GPQA Diamond | 53.6 | 75.4 |
| Humanity's Last Exam | — | 11.7 |
coding
| Benchmark | GPT-4o | Nemotron 3.5 Lightning (30B A3B) |
|---|---|---|
| SWE-bench Verified | — | 51.6 |
| SciCode | — | 32.6 |
agent
| Benchmark | GPT-4o | Nemotron 3.5 Lightning (30B A3B) |
|---|---|---|
| BrowseComp | — | 37.0 |
Category Winners
coding
15.6
reasoning
39.2
agent
8.6