| Benchmark | GPT-5.6 Sol | o1 |
|---|---|---|
| AA Agentic Index | 57.8 | 31.1 |
| AA Intelligence | 61 | 24 |
| AA-LCR | 77.7 | 63.3 |
| AA-Omniscience | 22 | -10.6 |
| Artificial Analysis Coding Index | 78.3 | 39.7 |
| critpt | 32.3 | 0.3 |
| gdpval | 61.1 | 11.5 |
| GPQA Diamond | 94.6 | 75.7 |
| HLE | 49.5 | 7.7 |
| IFBench | 72.7 | 70.3 |
| OmniScience Accuracy | 59.4 | 34.7 |
| OmniScience Non-Hallucination | 10.6 | 30.7 |
| scicode | 56.9 | 35.8 |
| simplebench | 64.8 | 41.7 |
| strongreject | 98.5 | 97 |
| Terminal-Bench Hard | 65.9 | 12.9 |
| τ²-Bench Telecom (AA run) | 85.1 | 62.6 |
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.