Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Claude Opus 4.1 | GPT-5 Codex |
|---|---|---|
| GPQA Diamond | 80.9 | — |
math
| Benchmark | Claude Opus 4.1 | GPT-5 Codex |
|---|---|---|
| AIME 2025 | 78 | — |
coding
| Benchmark | Claude Opus 4.1 | GPT-5 Codex |
|---|---|---|
| SWE-bench Verified | 74.5 | 74.5 |
| TerminalBench | 43.3 | — |
general
| Benchmark | Claude Opus 4.1 | GPT-5 Codex |
|---|---|---|
| MMMLU | 89.5 | — |
agent
| Benchmark | Claude Opus 4.1 | GPT-5 Codex |
|---|---|---|
| TAU-Bench Retail | 82.4 | — |
Category Winners
coding
75.2
math
47.9
reasoning
76.0
general
73.8
agent
86.7