Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| GPQA Diamond | 84.8 | 81 |
| Humanity's Last Exam | — | 17.2 |
math
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| AIME 2025 | 54.8 | 93.9 |
coding
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| SWE-bench Verified | 70.3 | 68 |
| TerminalBench | 35.2 | 40.5 |
multimodal
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| MMMU | 75 | — |
general
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| MMMLU | 86.1 | — |
agent
| Benchmark | Claude 3.7 Sonnet | GLM-4.6 |
|---|---|---|
| TAU-Bench Retail | 81.2 | — |
| BrowseComp | — | 45.1 |
Category Winners
coding
Z
GLM-4.663.7
math
Z
GLM-4.676.0
reasoning
81.8
general
61.4
multimodal
70.1
agent
84.7