Web GUI Interaction on WebForge
2AccuracyVanilla agent
Evaluation Results
| Method | Links | |
|---|---|---|
| Vanilla agentBackbone Model=GPT-5.4-mini2026.06 | 2 | |
| + visual memoryBackbone Model=GPT-5.4-mini2026.06 | 2 | |
| AGMem (Ours)Backbone Model=GPT-5.4-mini2026.06 | 2 |
| Method | Links | |
|---|---|---|
| Vanilla agentBackbone Model=GPT-5.4-mini2026.06 | 2 | |
| + visual memoryBackbone Model=GPT-5.4-mini2026.06 | 2 | |
| AGMem (Ours)Backbone Model=GPT-5.4-mini2026.06 | 2 |