Autonomous Driving on NAVSIM v2 (navtest)
98.9NCELF-VLA-8B
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ELF-VLA-8BMethod Category=VLA Methods w/ Textual Reasoning2026.06 | 98.9 | 98.1 | 99.4 | 99.8 | 88.5 | 98.4 | 96.9 | 98.3 | 87.2 | 87.1 | |
| LaST-VLA-8BMethod Category=VLA Methods w/ Visual Reasoning2026.06 | 98.7 | 97.9 | 99.2 | 99.7 | 90.3 | 98.2 | 96.6 | 98.3 | 86.3 | 87.1 | |
| WAM-Flow-1.5BMethod Category=VLA Methods w/o Explicit Reasoning2026.06 | 98.5 | 94.5 | 99.5 | 99.8 | 86.9 | 96.8 | 97.4 | 97.6 | 73.9 | 84.7 | |
| DriveVLA-W0-3BMethod Category=VLA Methods w/o Explicit Reasoning2026.06 | 98.5 | 99.1 | 98 | 99.7 | 86.4 | 98.1 | 93.2 | 97.9 | 58.9 | 86.1 | |
| Senna-2-3BMethod Category=VLA Methods w/ Textual Reasoning2026.06 | 98.5 | 97.8 | 99.5 | 99.8 | 88.1 | 97.5 | 97 | 98.6 | 88.4 | 86.6 | |
| RecogDrive-8BMethod Category=VLA Methods w/o Explicit Reasoning2026.06 | 98.3 | 95.2 | 99.5 | 99.8 | 87.1 | 97.5 | 96.6 | 98.3 | 86.5 | 83.6 | |
| DiffusionDriveMethod Category=Traditional End-to-End Methods2026.06 | 98.2 | 95.9 | 99.4 | 99.8 | 87.5 | 97.3 | 96.8 | 98.3 | 87.7 | 84.5 | |
| DriveSuprimMethod Category=Traditional End-to-End Methods2026.06 | 97.5 | 96.5 | 99.4 | 99.6 | 88.4 | 96.6 | 95.5 | 98.3 | 77 | 83.1 | |
| IRR-Drive-3BMethod Category=VLA Methods w/ Visual Reasoning, Model size=3B2026.06 | 97.3 | 97.4 | 98.7 | 99.6 | 91.9 | 97.1 | 95.7 | 97.9 | 76.9 | 87.9 | |
| HydraMDP++Method Category=Traditional End-to-End Methods2026.06 | 97.2 | 97.5 | 99.4 | 99.6 | 83.1 | 96.5 | 94.4 | 98.2 | 70.9 | 81.4 | |
| IRR-Drive-4BMethod Category=VLA Methods w/ Visual Reasoning, Model size=4B2026.06 | 97 | 98.3 | 98.9 | 99.5 | 92.3 | 96.8 | 95.8 | 97.6 | 82.2 | 89 |