Neura Intelligence Index
AI Model Benchmarks
| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro |
|---|---|---|
| GPQA Diamond | 87 | 94.3 |
| ARC-AGI v2 | 37.6 | 77.1 |
| Humanity's Last Exam | — | 51.4 |
| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro |
|---|---|---|
| SWE-bench Verified | 80.9 | 80.6 |
| SWE-Bench Pro | — | 54.2 |
| SciCode | — | 59 |
| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro |
|---|---|---|
| OSWorld | 66.3 | — |
| MCP Atlas | 62.3 | 69.2 |
| BrowseComp | — | 85.9 |
| APEX Agents | — | 33.5 |
| Benchmark | Claude Opus 4.5 | Gemini 3.1 Pro |
|---|---|---|
| MMMU-Pro | — | 80.5 |