Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | DeepSeek-V3.2 (Thinking) | Nemotron 3.5 Super VL (120B A12B) |
|---|---|---|
| GPQA Diamond | 82.4 | 89.5 |
| Humanity's Last Exam | 25.1 | 31.8 |
math
| Benchmark | DeepSeek-V3.2 (Thinking) | Nemotron 3.5 Super VL (120B A12B) |
|---|---|---|
| AIME 2025 | 93.1 | — |
coding
| Benchmark | DeepSeek-V3.2 (Thinking) | Nemotron 3.5 Super VL (120B A12B) |
|---|---|---|
| SWE-bench Verified | 73.1 | 71.1 |
| SWE-Bench Pro | — | 51.8 |
| SciCode | — | 47.1 |
agent
| Benchmark | DeepSeek-V3.2 (Thinking) | Nemotron 3.5 Super VL (120B A12B) |
|---|---|---|
| BrowseComp | 51.4 | 78 |
| Toolathlon | 35.2 | — |
multimodal
| Benchmark | DeepSeek-V3.2 (Thinking) | Nemotron 3.5 Super VL (120B A12B) |
|---|---|---|
| CharXiv Reasoning | — | 80.8 |
Category Winners
coding
67.2
math
74.9
reasoning
67.8
multimodal
65.8
agent
73.7