Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-5.1 Thinking | Grok-3 Mini |
|---|---|---|
| GPQA Diamond | 88.1 | 84 |
math
| Benchmark | GPT-5.1 Thinking | Grok-3 Mini |
|---|---|---|
| AIME 2025 | 94 | 90.8 |
| FrontierMath | 26.7 | — |
coding
| Benchmark | GPT-5.1 Thinking | Grok-3 Mini |
|---|---|---|
| SWE-bench Verified | 76.3 | — |
multimodal
| Benchmark | GPT-5.1 Thinking | Grok-3 Mini |
|---|---|---|
| MMMU | 85.4 | — |
Category Winners
coding
74.5
math
X
Grok-3 Mini71.1
reasoning
85.8
multimodal
89.6