| Benchmark | GLM-5.2 Full Open Source | Kimi K2 (Reasoning) |
|---|---|---|
| AA Agentic Index | 45.7 | 47.9 |
| AA Intelligence | 52.6 | 33.5 |
| AA-LCR | 76.7 | 70.3 |
| AA-Omniscience | 4.4 | -21.4 |
| Artificial Analysis Coding Index | 68.8 | 34.8 |
| critpt | 20.9 | 2.6 |
| gdpval | 50.3 | 24.5 |
| GPQA Diamond | 91.2 | 84.5 |
| HLE | 54.7 | 23.9 |
| HLE (with tools) | 54.7 | 44.9 |
| HMMT 2025 | 94.4 | 89.4 |
| hmmt_nov_2025 | 94.4 | 89.2 |
| IFBench | 73.3 | 68.1 |
| imo_answer_bench | 91 | 78.6 |
| OmniScience Accuracy | 24.3 | 30.9 |
| OmniScience Non-Hallucination | 73.7 | 25.8 |
| ResearchRubrics | 71.1 | 56.2 |
| scicode | 50.5 | 44.8 |
| simplebench | 58.8 | 39.6 |
| Terminal-Bench Hard | 50.8 | 31.1 |
| τ²-Bench Telecom (AA run) | 99.1 | 93 |
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.