Planning on GEODRIVE-BENCH
92.7Success Rate (CN)DRIVEOPD†
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DRIVEOPD†Setting=Direct, Backbone=Qwen2.5-VL-7B2026.06 | 92.7 | 88.8 | 75.4 | 84 | 83.4 | 92.1 | |
| Gemma3-12BSetting=Rule-Given2026.06 | 92.4 | 88.8 | 74.3 | 89.8 | 90.2 | 83.4 | |
| InternVL3.5-8BSetting=Rule-Given2026.06 | 91.1 | 90.4 | 75.9 | 83.4 | 82.8 | 80.1 | |
| DRIVEOPD‡Setting=Direct, Backbone=InternVL3-8B2026.06 | 90.8 | 89.4 | 79.7 | 89.3 | 84.7 | 91.4 | |
| DRIVEOPD‡Setting=Rule-Given, Backbone=InternVL3-8B2026.06 | 89.4 | 90.4 | 76.5 | 86.1 | 84.7 | 82.1 | |
| Gemma3-12BSetting=Direct2026.06 | 88.4 | 86.2 | 74.3 | 82.9 | 75.5 | 67.5 | |
| DRIVEOPD‡Setting=Reasoning, Backbone=InternVL3-8B2026.06 | 88.4 | 86.7 | 69.5 | 82.9 | 87.1 | 84.1 | |
| Gemma3-12BSetting=Reasoning2026.06 | 88.1 | 83.5 | 56.7 | 74.3 | 71.8 | 60.3 | |
| DRIVEOPD†Setting=Reasoning, Backbone=Qwen2.5-VL-7B2026.06 | 85.1 | 80.9 | 59.9 | 81.3 | 74.2 | 54.3 | |
| InternVL3-8BSetting=Rule-Given2026.06 | 83.5 | 88.3 | 75.9 | 81.8 | 82.2 | 72.2 | |
| Qwen3-VL-8BSetting=Rule-Given2026.06 | 83.2 | 86.7 | 76.5 | 87.7 | 85.9 | 79.5 | |
| DRIVEOPD†Setting=Rule-Given, Backbone=Qwen2.5-VL-7B2026.06 | 83.2 | 89.4 | 54.5 | 83.4 | 83.4 | 58.3 | |
| Qwen2.5-VL-7BSetting=Rule-Given2026.06 | 79.9 | 79.8 | 66.3 | 81.8 | 71.8 | 53.6 | |
| LLaVA-1.6-7BSetting=Direct2026.06 | 78.2 | 55.3 | 78.1 | 85 | 86.5 | 69.5 | |
| LLaVA-1.6-7BSetting=Rule-Given2026.06 | 76.6 | 82.4 | 66.3 | 79.1 | 73 | 55 | |
| Qwen2.5-VL-7BSetting=Direct2026.06 | 75.2 | 67 | 65.8 | 77.5 | 75.5 | 76.2 | |
| Qwen3-VL-8BSetting=Direct2026.06 | 74.3 | 72.9 | 63.1 | 84 | 65.6 | 57 | |
| Llama-3.2-11B-VSetting=Rule-Given2026.06 | 73.3 | 78.2 | 64.2 | 73.8 | 78.5 | 60.3 | |
| LLaVA-1.6-7BSetting=Reasoning2026.06 | 69.3 | 69.7 | 59.4 | 72.2 | 55.2 | 58.9 | |
| InternVL3-8BSetting=Direct2026.06 | 64.7 | 83 | 73.8 | 84.5 | 80.4 | 78.8 | |
| InternVL3-8BSetting=Reasoning2026.06 | 59.7 | 71.3 | 59.4 | 78.6 | 72.4 | 66.2 | |
| InternVL3.5-8BSetting=Direct2026.06 | 57.4 | 72.9 | 74.3 | 88.8 | 73 | 77.5 | |
| InternVL3.5-8BSetting=Reasoning2026.06 | 54.8 | 73.9 | 69 | 78.1 | 69.9 | 71.5 | |
| Qwen3-VL-8BSetting=Reasoning2026.06 | 53.5 | 74.5 | 35.8 | 78.1 | 73 | 39.1 | |
| Llama-3.2-11B-VSetting=Reasoning2026.06 | 43.2 | 64.4 | 68.4 | 72.2 | 66.3 | 72.2 | |
| Qwen2.5-VL-7BSetting=Reasoning2026.06 | 39.9 | 67.6 | 64.2 | 71.7 | 76.7 | 76.2 | |
| Llama-3.2-11B-VSetting=Direct2026.06 | 31 | 73.9 | 65.8 | 67.4 | 69.3 | 69.5 |