Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 3.1 Pro | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 94.3 | 93.6 |
| Humanity's Last Exam | 51.4 | 52.2 |
| ARC-AGI v2 | 77.1 | 85 |
coding
| Benchmark | Gemini 3.1 Pro | GPT-5.5 |
|---|---|---|
| SWE-bench Verified | 80.6 | — |
| SWE-Bench Pro | 54.2 | 58.6 |
| SciCode | 59 | — |
multimodal
| Benchmark | Gemini 3.1 Pro | GPT-5.5 |
|---|---|---|
| MMMU-Pro | 80.5 | 83.2 |
agent
| Benchmark | Gemini 3.1 Pro | GPT-5.5 |
|---|---|---|
| BrowseComp | 85.9 | 84.4 |
| MCP Atlas | 69.2 | 75.3 |
| APEX Agents | 33.5 | — |
| Toolathlon | — | 55.6 |
math
| Benchmark | Gemini 3.1 Pro | GPT-5.5 |
|---|---|---|
| FrontierMath | — | 35.4 |
Category Winners
coding
68.1
math
51.0
reasoning
91.0
general
90.3
multimodal
86.6
agent
73.6