Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| GPQA Diamond | 81 | 83.3 |
| Humanity's Last Exam | 17.7 | 14.7 |
| ARC-AGI v2 | — | 6.5 |
math
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| AIME 2025 | 87.5 | 86.4 |
| FrontierMath | — | 15.8 |
coding
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| SWE-bench Verified | 44.6 | 69.1 |
| TerminalBench | 5.7 | — |
general
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| SimpleQA | 92.3 | — |
agent
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| BrowseComp | 8.9 | 49.7 |
multimodal
| Benchmark | DeepSeek-R1-0528 | o3 |
|---|---|---|
| MMMU | — | 82.9 |
| MMMU-Pro | — | 76.4 |
| CharXiv Reasoning | — | 78.6 |
Category Winners
coding
59.1
math
65.5
reasoning
50.0
general
98.2
multimodal
75.1
agent
23.1