Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GLM-5.3-Flash | GPT-6.1 Sol |
|---|---|---|
| Humanity's Last Exam | 55.3 | — |
agent
| Benchmark | GLM-5.3-Flash | GPT-6.1 Sol |
|---|---|---|
| Toolathlon | 78.4 | — |
| APEX Agents | — | 60 |
multimodal
| Benchmark | GLM-5.3-Flash | GPT-6.1 Sol |
|---|---|---|
| CharXiv Reasoning | 89.4 | — |
| MMMU-Pro | — | 86.0 |
Category Winners
reasoning
Z
GLM-5.3-Flash90.5
multimodal
90.1
agent
98.3