| Benchmark | Claude 4 Opus | GLM-5.1 |
|---|---|---|
| AA Intelligence | 31.7 | 41 |
| AA-LCR | 40 | 68 |
| Artificial Analysis Coding Index | 34 | 55.8 |
| frontiermath_tier_4 | 4.2 | 12.5 |
| GPQA Diamond | 79.6 | 86.8 |
| HLE | 12.3 | 52.3 |
| HMMT 2025 | 15.9 | 94 |
| IFBench | 53.7 | 76.3 |
| livebench | 74.6 | 70.2 |
| MMLU-Pro | 86.6 | 86 |
| scicode | 40.9 | 43.8 |
| simplebench | 58.8 | 55.1 |
| Terminal-Bench Hard | 31.1 | 43.2 |
| τ²-Bench Telecom (AA run) | 73.4 | 97.7 |
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.