Step Execution on GUI-Odyssey
89.1Step Success Rate
UI-TARS-7B w/ GPT-4.1
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.03 | 89.1 | |
2026.03 | 88.6 | |
2026.03 | 88.2 | |
2026.03 | 87 | |
2026.03 | 83.4 | |
2026.03 | 62 | |
2026.03 | 62 | |
2026.03 | 56.4 | |
2026.03 | 54.5 | |
2026.03 | 50.9 | |
2026.03 | 41.3 | |
2026.03 | 38.8 | |
2026.03 | 5.4 | |
2026.03 | 3.1 |