Prediction on GEODRIVE-BENCH
75.8Performance (CN)Qwen3-VL-8B
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Qwen3-VL-8BSetting=Rule-Given2026.06 | 75.8 | 80.6 | 70.2 | 84.3 | 77.5 | 81.2 | |
| DRIVEOPD†Setting=Direct, Backbone=Qwen2.5-VL-7B2026.06 | 74.7 | 72.6 | 59 | 74.8 | 65.1 | 72.7 | |
| Gemma3-12BSetting=Rule-Given2026.06 | 73.1 | 73.7 | 55.3 | 83.3 | 70.4 | 77.8 | |
| DRIVEOPD‡Setting=Rule-Given, Backbone=InternVL3-8B2026.06 | 72 | 80.1 | 62.7 | 85.7 | 71 | 77.8 | |
| DRIVEOPD‡Setting=Reasoning, Backbone=InternVL3-8B2026.06 | 71.5 | 81.2 | 68.3 | 78.6 | 68 | 76.1 | |
| Qwen3-VL-8BSetting=Direct2026.06 | 71 | 62.9 | 69.6 | 49 | 72.2 | 78.4 | |
| InternVL3-8BSetting=Rule-Given2026.06 | 71 | 80.6 | 58.4 | 86.7 | 79.9 | 75 | |
| DRIVEOPD‡Setting=Rule-Given, Backbone=InternVL3-8B2026.06 | 71 | 78.1 | 66.7 | 59 | 56.7 | 71 | |
| Qwen2.5-VL-7BSetting=Rule-Given2026.06 | 70.4 | 72 | 64 | 80 | 68.6 | 64.8 | |
| DRIVEOPD†Setting=Rule-Given, Backbone=Qwen2.5-VL-7B2026.06 | 70.4 | 74.7 | 59 | 77.6 | 63.3 | 66.5 | |
| InternVL3.5-8BSetting=Direct2026.06 | 69.9 | 66.7 | 68.9 | 43.3 | 70.4 | 75 | |
| InternVL3.5-8BSetting=Rule-Given2026.06 | 69.9 | 74.7 | 65.8 | 84.8 | 71.6 | 76.1 | |
| DRIVEOPD‡Setting=Direct, Backbone=InternVL3-8B2026.06 | 69.9 | 78 | 67.1 | 81.4 | 60.9 | 83 | |
| DRIVEOPD‡Setting=Reasoning, Backbone=InternVL3-8B2026.06 | 69.5 | 73.8 | 63.8 | 50.5 | 52.9 | 70.5 | |
| Llama-3.2-11B-VSetting=Rule-Given2026.06 | 67.7 | 73.7 | 62.1 | 80 | 61.5 | 67 | |
| InternVL3-8BSetting=Direct2026.06 | 67.2 | 73.7 | 64 | 62.9 | 70.4 | 80.1 | |
| DRIVEOPD†Setting=Reasoning, Backbone=Qwen2.5-VL-7B2026.06 | 67.2 | 69.4 | 59 | 72.9 | 63.3 | 60.8 | |
| Qwen3-VL-8BSetting=Rule-Given2026.06 | 67.1 | 70 | 65.7 | 52.9 | 51.9 | 68.6 | |
| LLaVA-1.6-7BSetting=Rule-Given2026.06 | 65.6 | 74.2 | 60.2 | 81 | 67.5 | 59.1 | |
| DRIVEOPD‡Setting=Direct, Backbone=InternVL3-8B2026.06 | 64.8 | 68.6 | 65.2 | 48.6 | 51.9 | 61.9 | |
| InternVL3-8BSetting=Rule-Given2026.06 | 63.8 | 63.3 | 62.4 | 50.5 | 58.6 | 68.6 | |
| Qwen2.5-VL-7BSetting=Direct2026.06 | 62.9 | 66.7 | 54 | 45.2 | 56.2 | 67.6 | |
| Gemma3-12BSetting=Rule-Given2026.06 | 62.4 | 62.4 | 65.2 | 46.7 | 52.4 | 56.7 | |
| InternVL3.5-8BSetting=Rule-Given2026.06 | 60 | 75.2 | 67.1 | 51.9 | 49.5 | 67.6 | |
| Gemma3-12BSetting=Direct2026.06 | 59.7 | 60.2 | 54 | 41.4 | 55.6 | 73.3 | |
| InternVL3-8BSetting=Reasoning2026.06 | 59.7 | 69.4 | 61.5 | 46.2 | 70.4 | 75.6 | |
| DRIVEOPD†Setting=Rule-Given, Backbone=Qwen2.5-VL-7B2026.06 | 56.7 | 69 | 55.7 | 43.3 | 57.1 | 62.4 | |
| Qwen2.5-VL-7BSetting=Rule-Given2026.06 | 56.2 | 59.5 | 49.5 | 41.9 | 51 | 58.6 | |
| LLaVA-1.6-7BSetting=Reasoning2026.06 | 55.9 | 63.4 | 50.9 | 31.4 | 60.9 | 64.2 | |
| Gemma3-12BSetting=Reasoning2026.06 | 55.9 | 61.8 | 55.3 | 32.9 | 61.5 | 64.8 | |
| Llama-3.2-11B-VSetting=Reasoning2026.06 | 55.4 | 55.4 | 55.3 | 33.3 | 55.6 | 58.5 | |
| Qwen3-VL-8BSetting=Reasoning2026.06 | 54.8 | 53.8 | 56.5 | 40 | 65.7 | 69.9 | |
| DRIVEOPD†Setting=Direct, Backbone=Qwen2.5-VL-7B2026.06 | 54.3 | 67.6 | 54.8 | 28.1 | 51 | 60.5 | |
| Qwen2.5-VL-7BSetting=Reasoning2026.06 | 53.8 | 56.5 | 57.8 | 33.8 | 60.4 | 59.7 | |
| InternVL3.5-8BSetting=Reasoning2026.06 | 53.8 | 53.8 | 64.6 | 32.4 | 56.8 | 67.6 | |
| Gemma3-12BSetting=Direct2026.06 | 51.9 | 67.6 | 51 | 37.6 | 39.5 | 43.8 | |
| DRIVEOPD†Setting=Reasoning, Backbone=Qwen2.5-VL-7B2026.06 | 51.4 | 62.4 | 52.4 | 33.8 | 44.3 | 60 | |
| InternVL3-8BSetting=Reasoning2026.06 | 50 | 45.2 | 39 | 34.8 | 36.2 | 46.7 | |
| InternVL3-8BSetting=Direct2026.06 | 48.6 | 39 | 33.8 | 27.1 | 33.3 | 43.3 | |
| Qwen2.5-VL-7BSetting=Reasoning2026.06 | 44.8 | 35.2 | 29 | 21.9 | 33.3 | 43.8 | |
| Gemma3-12BSetting=Reasoning2026.06 | 44.8 | 60 | 56.7 | 27.6 | 34.8 | 51 | |
| Qwen3-VL-8BSetting=Direct2026.06 | 44.3 | 56.2 | 30.5 | 30 | 31 | 46.7 | |
| LLaVA-1.6-7BSetting=Direct2026.06 | 44.1 | 50 | 42.2 | 21.4 | 52.7 | 57.4 | |
| LLaVA-1.6-7BSetting=Rule-Given2026.06 | 43.3 | 59.5 | 43.3 | 49 | 38.6 | 46.7 | |
| Llama-3.2-11B-VSetting=Rule-Given2026.06 | 43.3 | 52.9 | 48.6 | 34.3 | 44.8 | 50 | |
| Qwen3-VL-8BSetting=Reasoning2026.06 | 42.9 | 34.8 | 29 | 31.9 | 29 | 34.3 | |
| InternVL3.5-8BSetting=Reasoning2026.06 | 42.4 | 52.9 | 41.4 | 28.1 | 23.3 | 46.7 | |
| Qwen2.5-VL-7BSetting=Direct2026.06 | 40.5 | 47.1 | 30 | 26.7 | 30.5 | 48.1 | |
| InternVL3.5-8BSetting=Direct2026.06 | 36.2 | 51 | 43.8 | 31.9 | 23.8 | 37.6 | |
| Llama-3.2-11B-VSetting=Direct2026.06 | 34.4 | 55.9 | 36.6 | 32.9 | 59.2 | 56.8 | |
| Llama-3.2-11B-VSetting=Reasoning2026.06 | 32.9 | 39 | 34.8 | 26.2 | 30 | 41 | |
| Llama-3.2-11B-VSetting=Direct2026.06 | 29 | 37.6 | 40 | 32.4 | 23.3 | 38.1 | |
| LLaVA-1.6-7BSetting=Reasoning2026.06 | 28.6 | 37.1 | 27.1 | 28.6 | 16.2 | 37.6 | |
| LLaVA-1.6-7BSetting=Direct2026.06 | 22.4 | 34.3 | 27.6 | 28.6 | 24.8 | 35.7 |