Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GLM-4.5-Air | GPT OSS 20B |
|---|---|---|
| GPQA Diamond | 75 | 71.5 |
| Humanity's Last Exam | 10.6 | 10.9 |
coding
| Benchmark | GLM-4.5-Air | GPT OSS 20B |
|---|---|---|
| SWE-bench Verified | 57.6 | — |
| TerminalBench | 30 | — |
| SciCode | 37.3 | — |
agent
| Benchmark | GLM-4.5-Air | GPT OSS 20B |
|---|---|---|
| BrowseComp | 21.3 | — |
| TAU-Bench Retail | 77.9 | 54.8 |
Category Winners
coding
Z
GLM-4.5-Air27.2
reasoning
Z
GLM-4.5-Air39.7
agent
Z
GLM-4.5-Air39.8