Planning on NAVSIM v1
94.8PDMSHuman Driver
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Human DriverVenue=NeurIPS’24, Category=Baselines & Oracle2026.05 | 94.8 | 100 | 100 | 87.5 | 100 | 99.9 | |
| ChainFlow-VLATraining Split=trainval2026.05 | 94.8 | 99.2 | 99 | 91.9 | 97.2 | 99.9 | |
| RAP-DINOVenue=ICLR’26, Category=End-to-End, pre-trained on private dataset=10x larger than default2026.05 | 93.8 | 99.1 | 98.9 | 90.3 | 96.7 | 100 | |
| DrivoRVenue=CVPR’26, Category=End-to-End, Training Split=trainval2026.05 | 93.7 | 99 | 98.9 | 90 | 96.7 | 100 | |
| ChainFlow-VLATraining Split=train2026.05 | 93.6 | 98.8 | 98.6 | 90.8 | 96.1 | 100 | |
| DriveSuprimVenue=AAAI’26, Category=End-to-End2026.05 | 93.5 | 98.6 | 98.6 | 91.3 | 95.5 | 100 | |
| DrivoRVenue=CVPR’26, Category=End-to-End, Training Split=train2026.05 | 93.1 | 98.9 | 98.3 | 89.1 | 96.2 | 100 | |
| CentaurVenue=arXiv’25, Category=End-to-End2026.05 | 92.6 | 99.5 | 98.9 | 85.9 | 98 | 100 | |
| LatentVLAVenue=arXiv’26, Category=VLA-based2026.05 | 92.4 | 98.9 | 98.2 | 88.2 | 96 | 100 | |
| iPadVenue=arXiv’25, Category=End-to-End2026.05 | 91.7 | 98.6 | 98.3 | 88 | 94.9 | 100 | |
| SGDriveVenue=CVPR’26, Category=VLA-based2026.05 | 91.1 | 98.6 | 97.8 | 85.8 | 96.2 | 100 | |
| Hydra-MDP++Venue=arXiv’25, Category=End-to-End2026.05 | 91 | 98.6 | 98.6 | 85.7 | 95.1 | 100 | |
| ReCogDriveVenue=ICLR’26, Category=VLA-based2026.05 | 90.8 | 97.9 | 97.3 | 87.3 | 94.9 | 100 | |
| SpanVLAVenue=arXiv’26, Category=VLA-based2026.05 | 90.3 | 99.1 | 97.1 | 86.3 | 95.2 | 100 | |
| DriveVLA-W0Venue=ICLR’26, Category=VLA-based2026.05 | 90.2 | 98.7 | 99.1 | 83.3 | 95.3 | 99.3 | |
| AutoVLAVenue=NeurIPS’25, Category=VLA-based2026.05 | 89.1 | 98.4 | 95.6 | 81.9 | 98 | 99.9 | |
| TrajDiffVenue=arXiv’25, Category=End-to-End, Training Split=trainval2026.05 | 88.5 | 98.1 | 97 | 82.7 | 94.3 | 100 | |
| DiffusionDriveVenue=CVPR’25, Category=End-to-End2026.05 | 88.1 | 98.2 | 96.2 | 82.2 | 94.7 | 100 | |
| TrajDiffVenue=arXiv’25, Category=End-to-End, Training Split=train2026.05 | 86.4 | 98 | 95 | 80.8 | 93.7 | 100 | |
| FSDriveVenue=NeurIPS’25, Category=VLA-based2026.05 | 85.1 | 98.2 | 93.8 | 80.1 | 93.3 | 99.9 | |
| UniADVenue=CVPR’23, Category=End-to-End2026.05 | 83.4 | 97.8 | 91.9 | 78.8 | 92.9 | 100 | |
| UniVLAVenue=RSS’25, Category=VLA-based2026.05 | 81.7 | 96.9 | 91.1 | 76.8 | 91.7 | 96.7 | |
| VADv2Venue=arXiv’24, Category=End-to-End2026.05 | 80.9 | 97.2 | 89.1 | 76 | 91.6 | 100 |