| Benchmark | Model | Rank | Results |
|---|---|---|---|
| agents-on-bench | GLM-5 | #1 | Accuracy: 89.7 |
| code-generation-on-humaneval | GLM 5 Thinking | #6 | Pass@1: 97.0 |
| agents-on-mcpatlas-public-500-tasks | GLM-5 | #8 | Accuracy: 67.8 |
| code-generation-on-swe-bench-verified | GLM-5 (OpenHands) | #15 | Accuracy: 77.8 |
| code-generation-on-swe-bench-multilingual | GLM-5 (OpenHands) | #16 | Accuracy: 73.3 |
| deep-research-agents-on-browsecomp | GLM-5 (w/ context manage) | #20 | Accuracy: 75.9 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | GLM-5 | #29 | Accuracy: 86.0 |