Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | ERNIE 5.0 | GLM-5V-Turbo |
|---|---|---|
| GPQA Diamond | 85 | — |
| Humanity's Last Exam | 39 | — |
math
| Benchmark | ERNIE 5.0 | GLM-5V-Turbo |
|---|---|---|
| AIME 2025 | 87 | — |
general
| Benchmark | ERNIE 5.0 | GLM-5V-Turbo |
|---|---|---|
| SimpleQA | 75 | — |
agent
| Benchmark | ERNIE 5.0 | GLM-5V-Turbo |
|---|---|---|
| OSWorld | — | 62.3 |
Category Winners
math
64.7
reasoning
74.6
general
92.2
agent
Z
GLM-5V-Turbo77.3