Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT OSS 120B High | Qwen3.5-397B-A17B |
|---|---|---|
| GPQA Diamond | 80.9 | 88.4 |
| Humanity's Last Exam | — | 28.7 |
math
| Benchmark | GPT OSS 120B High | Qwen3.5-397B-A17B |
|---|---|---|
| AIME 2025 | 92.5 | — |
general
| Benchmark | GPT OSS 120B High | Qwen3.5-397B-A17B |
|---|---|---|
| MMMLU | 83.8 | 88.5 |
coding
| Benchmark | GPT OSS 120B High | Qwen3.5-397B-A17B |
|---|---|---|
| SWE-bench Verified | — | 76.4 |
agent
| Benchmark | GPT OSS 120B High | Qwen3.5-397B-A17B |
|---|---|---|
| BrowseComp | — | 69 |
| Toolathlon | — | 38.3 |
Category Winners
coding
Q
Qwen3.5-397B-A17B74.3
math
74.3
reasoning
73.4
general
Q
Qwen3.5-397B-A17B70.8
agent
Q
Qwen3.5-397B-A17B35.4