Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | Claude Haiku 5.5 | Qwen3.7 Max |
|---|---|---|
| SWE-Bench Pro | 64.8 | 60.6 |
| SWE-bench Verified | — | 80.4 |
| SciCode | — | 53.5 |
reasoning
| Benchmark | Claude Haiku 5.5 | Qwen3.7 Max |
|---|---|---|
| GPQA Diamond | — | 92.4 |
| Humanity's Last Exam | — | 41.4 |
general
| Benchmark | Claude Haiku 5.5 | Qwen3.7 Max |
|---|---|---|
| MMMLU | — | 90.3 |
agent
| Benchmark | Claude Haiku 5.5 | Qwen3.7 Max |
|---|---|---|
| MCP Atlas | — | 76.4 |
Category Winners
coding
75.2
reasoning
Q
Qwen3.7 Max79.6
general
Q
Qwen3.7 Max76.9
agent
Q
Qwen3.7 Max67.4