Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 3.1 Pro | Hy3 |
|---|---|---|
| GPQA Diamond | 94.3 | 90.4 |
| Humanity's Last Exam | 51.4 | — |
| ARC-AGI v2 | 77.1 | — |
coding
| Benchmark | Gemini 3.1 Pro | Hy3 |
|---|---|---|
| SWE-bench Verified | 80.6 | 78 |
| SWE-Bench Pro | 54.2 | 57.9 |
| SciCode | 59 | — |
multimodal
| Benchmark | Gemini 3.1 Pro | Hy3 |
|---|---|---|
| MMMU-Pro | 80.5 | — |
agent
| Benchmark | Gemini 3.1 Pro | Hy3 |
|---|---|---|
| BrowseComp | 85.9 | 84.2 |
| MCP Atlas | 69.2 | 79.1 |
| APEX Agents | 33.5 | 25.6 |
| Toolathlon | — | 48.5 |
Category Winners
coding
71.2
reasoning
87.9
general
51.8
multimodal
81.9
agent
58.9