| agents-on-agents-last-exam-ale-cli | GLM-5.3 (Claude Code) | #3 | Pass Rate: 28.5 |
| agents-on-posttrainbench-v1-1 | GLM-5.3 (Claude Code) | #3 | Accuracy: 39.8 |
| code-generation-on-terminal-bench-2-1 | GLM-5.3 (Claude Code) | #4 | Accuracy: 88.2 |
| exploit-generation-on-exploitbench | GLM-5.3 (Claude Code) | #4 | Coverage: 0.544 |
| vulnerability-reproduction-on-cybergym | GLM-5.3 (Claude Code) | #4 | Success Rate: 84.5 |
| code-generation-on-nl2repo | GLM-5.3 | #5 | Accuracy: 58.0 |
| code-generation-on-terminal-bench-3-0 | GLM-5.3 (Claude Code) | #5 | Accuracy: 28.3 |
| agents-on-automationbench-public | GLM-5.3 | #6 | Success Rate: 48.2 |
| agents-on-toolathlon-verified-108-tasks | GLM-5.3 | #9 | Pass@1: 0.73 |
| code-generation-on-programbench | GLM-5.3 | #10 | Almost Resolved: 19.0 |
| code-generation-on-deepswe | GLM-5.3 (MiniSWE-Agent) | #13 | Pass@1: 0.669 |