Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 91.9 | 92.4 |
| Humanity's Last Exam | 45.8 | 34.5 |
| ARC-AGI v2 | 31.1 | 52.9 |
math
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| AIME 2025 | 100 | 100 |
| FrontierMath | — | 40.3 |
coding
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| SWE-bench Verified | 76.2 | 80 |
multimodal
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| MMMU-Pro | 81 | 79.5 |
| CharXiv Reasoning | 81.4 | 82.1 |
| ScreenSpot Pro | 72.7 | 86.3 |
agent
| Benchmark | Gemini 3 Pro | GPT-5.2 |
|---|---|---|
| BrowseComp | — | 65.8 |
| Toolathlon | — | 46.3 |
| MCP Atlas | — | 60.6 |
Category Winners
coding
81.1
math
84.3
reasoning
68.4
general
74.6
multimodal
80.3
agent
34.1