Navigation on CityNav New York 1.0
95Success RateAgentNav (O3)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AgentNav (O3)Agent Config=AgentNav, Backbone=O32025.12 | 95 | 0.759 | 84.93 | |
| AgentNav (GPT 5)Agent Config=AgentNav, Backbone=GPT 52025.12 | 94 | 0.711 | 82.98 | |
| AgentNav (GPT 4.1)Agent Config=AgentNav, Backbone=GPT 4.12025.12 | 92 | 0.557 | 75.27 | |
| AgentNav (GPT 4o)Agent Config=AgentNav, Backbone=GPT 4o2025.12 | 88 | 0.539 | 72.91 | |
| AgentNav (Gemini 2.5 Flash)Agent Config=AgentNav, Backbone=Gemini 2.5 Flash2025.12 | 73 | 0.471 | 74.75 | |
| GPT 5Agent Config=Base2025.12 | 54 | 0.375 | 55.97 | |
| O3Agent Config=Base2025.12 | 48 | 0.49 | 64.36 | |
| AgentNav (Qwen 2.5 VL 32b)Agent Config=AgentNav, Backbone=Qwen 2.5 VL 32b2025.12 | 32 | 0.153 | 56.39 | |
| GPT 4.1Agent Config=Base2025.12 | 15 | 0.097 | 42.27 | |
| GPT 4oAgent Config=Base2025.12 | 13 | 0.064 | 39.04 | |
| Gemini 2.5 FlashAgent Config=Base2025.12 | 12 | 0.06 | 41.57 | |
| Qwen 2.5 VL 32bAgent Config=Base2025.12 | 7 | 0.089 | 35.11 |