Collaborative Execution on Agentization Bench (Overall)
71.5Original Success Rate (Or.SR)Claude Code
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude Code2026.04 | 71.5 | 39.2 | |
| EnvX2026.04 | 69.4 | 41.4 | |
| Codex2026.04 | 67.2 | 40.3 | |
| OpenHands2026.04 | 65.1 | 46.2 |
| Method | Links | ||
|---|---|---|---|
| Claude Code2026.04 | 71.5 | 39.2 | |
| EnvX2026.04 | 69.4 | 41.4 | |
| Codex2026.04 | 67.2 | 40.3 | |
| OpenHands2026.04 | 65.1 | 46.2 |