| Benchmark | Gemini 1.5 Pro | GLM-5.1 |
|---|---|---|
| AA Intelligence | 10 | 41 |
| Artificial Analysis Coding Index | 23.6 | 55.8 |
| GPQA Diamond | 59.1 | 86.8 |
| HLE | 4.9 | 52.3 |
| MMLU-Pro | 75.8 | 86 |
| scicode | 29.5 | 43.8 |
| simplebench | 27.1 | 55.1 |
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.