Autonomous Driving on NAVSIM v1 (test)
100NCHuman driver
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Human driverModality=N/A2026.01 | 100 | 100 | 100 | 99.9 | 87.5 | 94.8 | |
| Human driver2026.01 | 100 | 100 | 100 | 99.9 | 87.5 | 94.8 | |
| Human Driver(Ground Truth)Type=Ground Truth2026.05 | 100 | 100 | 100 | 99.9 | 87.5 | 94.8 | |
| CentaurModality=Camera, Strategy=Test-time training2026.01 | 99.5 | 98.9 | 98 | 100 | 85.9 | 92.6 | |
| Centaur2026.01 | 99.5 | 98.9 | 98 | 100 | 85.9 | 92.6 | |
| TransDiffuserModality=Camera + LiDAR, Strategy=Best-of-N2026.01 | 99.4 | 96.5 | 97.8 | 99.4 | 94.1 | 94.9 | |
| DriveVLA-W0Modality=Camera, Strategy=Best-of-N2026.01 | 99.3 | 97.4 | 97 | 99.9 | 88.3 | 93 | |
| AutoVLAModality=Camera, Strategy=Best-of-N2026.01 | 99.1 | 97.1 | 97.1 | 100 | 87.6 | 92.1 | |
| RAP-DINOModality=Camera, Strategy=10x more training data2026.01 | 99.1 | 98.9 | 96.7 | 100 | 90.3 | 93.8 | |
| DrivoR (+65k SimScale data)Modality=Camera, Training Data=trainval + 65k SimScale2026.01 | 99.1 | 99 | 96.9 | 100 | 90.3 | 94 | |
| DrivoR (+134k SimScale data)Modality=Camera, Training Data=trainval + 134k SimScale2026.01 | 99.1 | 99.2 | 96.9 | 100 | 91.6 | 94.6 | |
| RAP-DINOTraining Data=10x larger than navtrain2026.01 | 99.1 | 98.9 | 96.7 | 100 | 90.3 | 93.8 | |
| DrivoRTraining Data=trainval, Data Augmentation (SimScale)=65k2026.01 | 99.1 | 99 | 96.9 | 100 | 90.3 | 94 | |
| DrivoRTraining Data=trainval, Data Augmentation (SimScale)=134k2026.01 | 99.1 | 99.2 | 96.9 | 100 | 91.6 | 94.6 | |
| WAM-diff2026.02 | 99.1 | 98.3 | 96.5 | 99.9 | 84.4 | 91 | |
| SpanVLACam.=true, Lid.=false, Evaluation Protocol=Post-RFT2026.04 | 99.1 | 97.1 | 95.2 | 100 | 86.3 | 90.3 | |
| RAP2026.05 | 99.1 | 98.9 | 96.7 | 100 | 90.3 | 93.8 | |
| NTR2026.05 | 99.1 | 98.8 | 96.9 | 100 | 90.8 | 94.1 | |
| DrivoR (trainval)Modality=Camera, Training Data=trainval2026.01 | 99 | 98.9 | 96.7 | 100 | 90 | 93.7 | |
| DrivoRTraining Data=trainval2026.01 | 99 | 98.9 | 96.7 | 100 | 90 | 93.7 | |
| ResAD (V2-99)Modality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.9 | 97.8 | 94.9 | 100 | 87 | 90.6 | |
| DrivingGPTModality=Camera, Strategy=Camera-only2026.01 | 98.9 | 90.7 | 94.9 | 95.6 | 79.7 | 82.4 | |
| DrivoR (train)Modality=Camera, Training Data=train2026.01 | 98.9 | 98.3 | 96.2 | 100 | 89.1 | 93.1 | |
| DrivingGPT2026.01 | 98.9 | 90.7 | 94.9 | 95.6 | 79.7 | 82.4 | |
| DrivoRTraining Data=train2026.01 | 98.9 | 98.3 | 96.2 | 100 | 89.1 | 93.1 | |
| DrivingGPT2026.02 | 98.9 | 90.7 | 94.9 | 95.6 | 79.7 | 82.4 | |
| DrivingGPT2026.05 | 98.9 | 90.7 | 94.9 | 95.6 | 79.7 | 82.4 | |
| DrivoR2026.05 | 98.9 | 98.3 | 96.2 | 100 | 89.1 | 93.7 | |
| SeerDrive (V2-99)Modality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.8 | 98.6 | 95.8 | 100 | 84.2 | 90.7 | |
| Hydra-MDP-C (V2-99)Modality=Camera + LiDAR, Strategy=Ensemble2026.01 | 98.7 | 98.2 | 95 | 100 | 86.5 | 91 | |
| DriveVLA-W0Modality=Camera, Strategy=Camera-only (General)2026.01 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| DriveVLA-W02026.01 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| DriveVLA-W02026.02 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| DriveVLA-W0Venue=NeurIPS 2025, Input=C&L2026.03 | 98.7 | 96.2 | 95.5 | 100 | 82.2 | 88.4 | |
| DriveVLA-W0Cam.=true, Lid.=false2026.04 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| DriveVLA-W0Variant=W02026.05 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| Hydra-MDP++Modality=Camera, Strategy=Camera-only2026.01 | 98.6 | 98.6 | 95.1 | 100 | 85.7 | 91 | |
| iPadModality=Camera, Strategy=Camera-only2026.01 | 98.6 | 98.3 | 94.9 | 100 | 88 | 91.7 | |
| DriveSuprimModality=Camera, Strategy=Camera-only2026.01 | 98.6 | 98.6 | 95.5 | 100 | 91.3 | 93.5 | |
| Hydra-MDP++2026.01 | 98.6 | 98.6 | 95.1 | 100 | 85.7 | 91 | |
| iPad2026.01 | 98.6 | 98.3 | 94.9 | 100 | 88 | 91.7 | |
| DriveSuprim2026.01 | 98.6 | 98.6 | 95.5 | 100 | 91.3 | 93.5 | |
| iPad2026.02 | 98.6 | 98.3 | 94.9 | 100 | 88 | 91.7 | |
| HybridDriveVLA2026.02 | 98.6 | 98.6 | 96.2 | 100 | 87.3 | 92.1 | |
| WOTEModality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.5 | 96.8 | 94.9 | 99.9 | 81.9 | 88.3 | |
| DIVERModality=Camera, Strategy=Camera-only2026.01 | 98.5 | 96.5 | 94.9 | 100 | 82.6 | 88.3 | |
| DIVER2026.01 | 98.5 | 96.5 | 94.9 | 100 | 82.6 | 88.3 | |
| WoTEImg. Backbone=ResNet-342025.12 | 98.5 | 96.8 | 94.9 | 99.9 | 81.9 | 88.3 | |
| DIVERImg. Backbone=ResNet-342025.12 | 98.5 | 96.5 | 94.9 | 100 | 82.6 | 88.3 | |
| WoTE2026.02 | 98.5 | 96.8 | 94.9 | 99.9 | 81.9 | 88.3 | |
| WOTEVenue=ICCV 2025, Input=C&L2026.03 | 98.5 | 96.8 | 94.9 | 99.9 | 81.9 | 88.3 | |
| WoTECam.=true, Lid.=true2026.04 | 98.5 | 96.8 | 94.4 | 99.9 | 81.9 | 88.3 | |
| DiffusionDrive (SimScale-planner)Backbone=DiffusionDrive, Settings=SimScale-planner, Real data amount=85K, Synthetic data amount=237K2026.05 | 98.5 | 97.1 | 94.7 | 100 | 83.1 | 88.9 | |
| DiffusionDrive (AutoScale)Backbone=DiffusionDrive, Settings=AutoScale, Real data amount=85K, Synthetic data amount=50K2026.05 | 98.5 | 96.7 | 94.8 | 100 | 82.9 | 88.7 | |
| DiffusionDrive (AutoScale)Backbone=DiffusionDrive, Settings=AutoScale, Real data amount=85K, Synthetic data amount=100K2026.05 | 98.5 | 96.9 | 95 | 100 | 83.1 | 89 | |
| GoalFlow (V2-99)Modality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.4 | 98.3 | 94.6 | 100 | 85 | 90.3 | |
| AutoVLAModality=Camera, Strategy=Camera-only (General)2026.01 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| AutoVLA2026.01 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| Hydra-MDPImg. Backbone=V2-992025.12 | 98.4 | 97.8 | 93.9 | 100 | 86.5 | 90.3 | |
| GoalFlowImg. Backbone=V2-992025.12 | 98.4 | 98.3 | 94.6 | 100 | 85 | 90.3 | |
| AutoVLA2026.02 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| AutoVLAVenue=NeurIPS 2025, Input=C-Only2026.03 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| AutoVLACam.=true, Lid.=false2026.04 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| DiffusionDrive (SimScale-recovery)Backbone=DiffusionDrive, Settings=SimScale-recovery, Real data amount=85K, Synthetic data amount=147K2026.05 | 98.4 | 96.5 | 94.7 | 100 | 82.8 | 88.4 | |
| AutoVLA2026.05 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| GoalFlow2026.05 | 98.4 | 98.3 | 94.6 | 100 | 85 | 90.3 | |
| Hydra-MDP*Img. Backbone=ResNet-342025.12 | 98.3 | 96 | 94.6 | 100 | 78.7 | 86.5 | |
| GoalFlow*Img. Backbone=ResNet-342025.12 | 98.3 | 93.8 | 94.3 | 100 | 79.8 | 85.7 | |
| ARTEMISImg. Backbone=ResNet-342025.12 | 98.3 | 95.1 | 94.3 | 100 | 81.4 | 87 | |
| DiffusionDriveV2Img. Backbone=ResNet-342025.12 | 98.3 | 97.9 | 94.8 | 99.9 | 87.5 | 91.2 | |
| Hydra-MDP2026.02 | 98.3 | 96 | 94.6 | 100 | 78.7 | 86.5 | |
| DiffusionDriveV22026.02 | 98.3 | 97.9 | 94.8 | 99.9 | 88 | 91.2 | |
| Hydra-MDPVenue=arXiv 2024, Input=C&L2026.03 | 98.3 | 96 | 94.6 | 100 | 78.7 | 86.5 | |
| Hydra-MDPCam.=true, Lid.=true2026.04 | 98.3 | 96 | 94.6 | 100 | 78.7 | 86.5 | |
| TransFuser (SimScale-planner)Backbone=TransFuser, Settings=SimScale-planner, Real data amount=85K, Synthetic data amount=237K2026.05 | 98.3 | 95.6 | 94.6 | 100 | 91.3 | 87.3 | |
| DiffusionDriveModality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DiffusionDriveModality=Camera, Strategy=Camera-only2026.01 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DiffusionDrive2026.01 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DiffusionDriveImg. Backbone=ResNet-342025.12 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DiffusionDrive2026.02 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DRAMACam.=true, Lid.=true2026.04 | 98.2 | 95.2 | 94.2 | 100 | 81.3 | 86.9 | |
| DiffusionDriveCam.=true, Lid.=true2026.04 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| ReCogDriveCam.=true, Lid.=false2026.04 | 98.2 | 97.8 | 95.2 | 99.8 | 83.5 | 89.6 | |
| TransFuser (SimScale-recovery)Backbone=TransFuser, Settings=SimScale-recovery, Real data amount=85K, Synthetic data amount=147K2026.05 | 98.2 | 95 | 94.3 | 100 | 80.5 | 86.4 | |
| DiffusionDrive (Real Data)Backbone=DiffusionDrive, Settings=Real Data, Real data amount=85K, Synthetic data amount=-2026.05 | 98.2 | 95.2 | 94.4 | 100 | 81.5 | 87.1 | |
| DiffusionDrive2026.05 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| DistillDriveModality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98.1 | 94.6 | 93.6 | 100 | 81 | 86.2 | |
| VAD-v2Modality=Camera, Strategy=Camera-only2026.01 | 98.1 | 94.8 | 94.3 | 100 | 80.6 | 86.2 | |
| PRIXModality=Camera, Strategy=Camera-only2026.01 | 98.1 | 96.3 | 94.1 | 100 | 82.3 | 87.8 | |
| VAD-v22026.01 | 98.1 | 94.8 | 94.3 | 100 | 80.6 | 86.2 | |
| PRIX2026.01 | 98.1 | 96.3 | 94.1 | 100 | 82.3 | 87.8 | |
| ImagiDrive2026.02 | 98.1 | 96.2 | 94.5 | 100 | 80.5 | 86.9 | |
| VAD-v22026.05 | 98.1 | 94.8 | 94.3 | 100 | 80.6 | 86.2 | |
| Hydra-MDP (V2-99)Modality=Camera + LiDAR, Strategy=Multi-modal2026.01 | 98 | 97.8 | 93.9 | 100 | 86.5 | 90.3 | |
| DRAMAModality=Camera, Strategy=Camera-only2026.01 | 98 | 93.1 | 94.8 | 100 | 80.1 | 85.5 | |
| DRAMA2026.01 | 98 | 93.1 | 94.8 | 100 | 80.1 | 85.5 | |
| DRAMAImg. Backbone=ResNet-342025.12 | 98 | 93.1 | 94.8 | 100 | 80.1 | 85.5 | |
| DRAMA2026.02 | 98 | 93.1 | 94.8 | 100 | 80.1 | 85.5 | |
| DRAMAVenue=arXiv 2024, Input=C&L2026.03 | 98 | 93.1 | 94.8 | 100 | 80.1 | 85.5 | |
| DreamerADVenue=-, Input=C-Only2026.03 | 98 | 97.2 | 94.3 | 100 | 83.1 | 88.7 |