VillagerAgent Escape Room Challenge
73.29Completion Rate (%)VillagerAgent (gpt-4-1106-preview)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VillagerAgent (gpt-4-1106-preview)Backbone=gpt-4-1106-preview2024.06 | 73.29 | 149.4 | 90.03 | |
| VillagerAgent (gpt-4-1106-preview, 3-agents)Backbone=gpt-4-1106-preview, Number of agents=32024.06 | 69.78 | 227.4 | 67.01 | |
| VillagerAgent (gemini-pro)Backbone=gemini-pro2024.06 | 69.2 | 153.3 | 80.35 | |
| VillagerAgent (glm-4)Backbone=glm-42024.06 | 68.17 | 100.8 | 95.3 |