Neura Intelligence Index
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| GPQA Diamond | 82.7 | 74.4 |
| Humanity's Last Exam | 22.8 | — |
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| AIME 2025 | 90.2 | 83.1 |
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| SWE-bench Verified | 53.7 | — |
| TerminalBench | 25.8 | — |
| SciCode | 42.0 | — |
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| BrowseComp | 31.3 | — |
| OSWorld | — | 30.6 |
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| MMMU-Pro | — | 63 |
| CharXiv Reasoning | — | 56.6 |
| ScreenSpot Pro | — | 57.3 |
| Benchmark | Nemotron 3 Super (120B A12B) | Qwen3 VL 30B A3B Thinking |
|---|---|---|
| SimpleQA | — | 23.9 |