| Benchmark | GPT-4o mini | Qwen3.6 Plus |
|---|---|---|
| AA Agentic Index | 1 | 29 |
| AA Intelligence | 7 | 40.5 |
| AI2D | 75.2 | 94.4 |
| Artificial Analysis Coding Index | 11.4 | 54.5 |
| gdpval | 0 | 32 |
| GPQA Diamond | 42.6 | 90.4 |
| HLE | 4.2 | 28.8 |
| IFBench | 31 | 75.2 |
| MMMU | 59.4 | 86 |
| MMMU-Pro | 41.5 | 78.8 |
| scicode | 22.9 | 40.7 |
| SWE-bench Verified | 8.7 | 78.8 |
| TauBench V3 - Banking | 2.9 | 20.8 |
| Terminal-Bench 2.1 | 5.6 | 61.4 |
| Video-MME | 68.9 | 84.2 |
| τ³-Bench | 2.9 | 70.7 |
Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.