Agent Workflow Execution on CUA-World-Long (test)
55.5Average Score
GPT-5.4
Updated 3mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 55.5 | 27.5 | |
2026.04 | 38.7 | 11.5 | |
2026.04 | 36.2 | 7.5 | |
2026.04 | 33.9 | 5.5 | |
2026.04 | 22.7 | 3 | |
2026.04 | 20.5 | 6 |
GPT-5.4
| Method | Links | ||
|---|---|---|---|
2026.04 | 55.5 | 27.5 | |
2026.04 | 38.7 | 11.5 | |
2026.04 | 36.2 | 7.5 | |
2026.04 | 33.9 | 5.5 | |
2026.04 | 22.7 | 3 | |
2026.04 | 20.5 | 6 |