Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | ERNIE 5.0 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 85 | 88.1 |
| Humanity's Last Exam | 39 | — |
math
| Benchmark | ERNIE 5.0 | GPT-5.1 |
|---|---|---|
| AIME 2025 | 87 | 94 |
| FrontierMath | — | 26.7 |
general
| Benchmark | ERNIE 5.0 | GPT-5.1 |
|---|---|---|
| SimpleQA | 75 | — |
coding
| Benchmark | ERNIE 5.0 | GPT-5.1 |
|---|---|---|
| SWE-bench Verified | — | 76.3 |
multimodal
| Benchmark | ERNIE 5.0 | GPT-5.1 |
|---|---|---|
| MMMU | — | 85.4 |
Category Winners
coding
74.5
math
64.7
reasoning
85.8
general
92.2
multimodal
89.6