Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | Claude Opus 4 | GPT-5.1 Thinking |
|---|---|---|
| HumanEval | 95.1 | — |
| SWE-bench Verified | 55.8 | 76.3 |
math
| Benchmark | Claude Opus 4 | GPT-5.1 Thinking |
|---|---|---|
| MATH | 85.2 | — |
| AIME 2024 | 35 | — |
| GSM8K | 97.8 | — |
| AIME 2025 | — | 94 |
| FrontierMath |
reasoning
| Benchmark | Claude Opus 4 | GPT-5.1 Thinking |
|---|---|---|
| GPQA Diamond | 74.8 | 88.1 |
| BigBench-Hard | 93.2 | — |
general
| Benchmark | Claude Opus 4 | GPT-5.1 Thinking |
|---|---|---|
| IFEval | 91.5 | — |
| MMLU-Pro | 83.5 | — |
| Chatbot Arena Elo | 1365 | — |
multimodal
| Benchmark | Claude Opus 4 | GPT-5.1 Thinking |
|---|---|---|
| MMMU | 74.2 | 85.4 |
Category Winners
coding
74.5
math
75.5
reasoning
85.8
general
96.0
multimodal
89.6