Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | GPT-5 Codex | Qwen3.7-Plus |
|---|---|---|
| SWE-bench Verified | 74.5 | 77.7 |
| SWE-Bench Pro | — | 57.6 |
| SciCode | — | 51.3 |
reasoning
| Benchmark | GPT-5 Codex | Qwen3.7-Plus |
|---|---|---|
| GPQA Diamond | — | 90.3 |
| Humanity's Last Exam | — | 34.7 |
multimodal
| Benchmark | GPT-5 Codex | Qwen3.7-Plus |
|---|---|---|
| MMMU-Pro | — | 79 |
| CharXiv Reasoning | — | 85.9 |
| ScreenSpot Pro | — | 79 |
general
| Benchmark | GPT-5 Codex | Qwen3.7-Plus |
|---|---|---|
| MMMLU | — | 89 |
agent
| Benchmark | GPT-5 Codex | Qwen3.7-Plus |
|---|---|---|
| MCP Atlas | — | 73.2 |
Category Winners
coding
71.0
reasoning
Q
Qwen3.7-Plus73.8
general
Q
Qwen3.7-Plus72.2
multimodal
Q
Qwen3.7-Plus82.7
agent
Q
Qwen3.7-Plus59.5