Overall on GEODRIVE-BENCH
78.6Overall ScoreDRIVEOPD‡
Evaluation Results
| Method | Links | |
|---|---|---|
| DRIVEOPD‡Setting=Rule-Given, Backbone=InternVL3-8B2026.06 | 78.6 | |
| Qwen3-VL-8BSetting=Rule-Given2026.06 | 77.7 | |
| DRIVEOPD‡Setting=Direct, Backbone=InternVL3-8B2026.06 | 77.3 | |
| InternVL3-8BSetting=Rule-Given2026.06 | 77.2 | |
| DRIVEOPD‡Setting=Reasoning, Backbone=InternVL3-8B2026.06 | 77.1 | |
| InternVL3.5-8BSetting=Rule-Given2026.06 | 76.8 | |
| DRIVEOPD†Setting=Direct, Backbone=Qwen2.5-VL-7B2026.06 | 75.7 | |
| Gemma3-12BSetting=Rule-Given2026.06 | 72.2 | |
| DRIVEOPD†Setting=Rule-Given, Backbone=Qwen2.5-VL-7B2026.06 | 71.6 | |
| Qwen2.5-VL-7BSetting=Rule-Given2026.06 | 70.3 | |
| DRIVEOPD†Setting=Reasoning, Backbone=Qwen2.5-VL-7B2026.06 | 68.4 | |
| Llama-3.2-11B-VSetting=Rule-Given2026.06 | 63.6 | |
| InternVL3-8BSetting=Direct2026.06 | 61 | |
| Gemma3-12BSetting=Direct2026.06 | 60.1 | |
| InternVL3.5-8BSetting=Direct2026.06 | 59.9 | |
| Qwen3-VL-8BSetting=Direct2026.06 | 58.8 | |
| Gemma3-12BSetting=Reasoning2026.06 | 58.5 | |
| Qwen2.5-VL-7BSetting=Direct2026.06 | 57.9 | |
| InternVL3-8BSetting=Reasoning2026.06 | 57.4 | |
| LLaVA-1.6-7BSetting=Rule-Given2026.06 | 57.1 | |
| InternVL3.5-8BSetting=Reasoning2026.06 | 54.6 | |
| Qwen2.5-VL-7BSetting=Reasoning2026.06 | 51.8 | |
| Qwen3-VL-8BSetting=Reasoning2026.06 | 50.5 | |
| Llama-3.2-11B-VSetting=Reasoning2026.06 | 47 | |
| Llama-3.2-11B-VSetting=Direct2026.06 | 45.5 | |
| LLaVA-1.6-7BSetting=Reasoning2026.06 | 43.2 | |
| LLaVA-1.6-7BSetting=Direct2026.06 | 42.2 |