Autonomous Driving Planning on NAVSIM v2 (Navtest)
100NCHuman
Evaluation results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HumanPerception=false2026.05 | 100 | 100 | 87.4 | 100 | 98.1 | 100 | 99.8 | 100 | 90.1 | — | — | — | 94.5 | — | |
| Human AgentSensors=-2026.06 | 100 | 100 | 87.4 | 100 | 98.1 | 100 | 99.8 | 100 | 90.1 | — | — | — | 90.3 | — | |
| Clover2026.05 | 99.4 | 99.3 | 86.9 | 98.9 | 98.2 | — | 99.5 | 95.2 | 75.5 | — | — | 99.8 | 90.4 | 87.2 | |
| AutoDrive-P3Perception=true2026.05 | 99.1 | 97.4 | 88 | 98.7 | 98.3 | 99.8 | 99.2 | 96.3 | 85.5 | — | — | — | 89.9 | — | |
| VegaSensors=1xC, Planner Type=VLA-based Planner2026.06 | 98.9 | 95.3 | 87 | 98.4 | 98.3 | 99.9 | 99.4 | 96.1 | 76.3 | — | — | — | 86.9 | — | |
| SpanVLAEvaluation protocol=Post-RFT2026.04 | 98.8 | 96.7 | 86.3 | 97.7 | 98.3 | 99.8 | 99.2 | 95 | 85.1 | 86.4 | — | — | — | — | |
| PWMPerception=false2026.05 | 98.8 | 95.9 | 86.4 | 98.4 | 98.3 | 99.9 | 99.4 | 97.6 | 85.3 | — | — | — | 88.2 | — | |
| Drive-JEPASensors=1xC, Planner Type=Traditional E2E-based Planner2026.06 | 98.8 | 97.4 | 83.5 | 98 | 98.1 | 99.8 | 99 | 96.2 | 85.6 | — | — | — | 85.4 | — | |
| DriveLaWPerception=false2026.05 | 98.7 | 96.9 | 87.5 | 98.3 | 98.4 | 99.8 | 99.6 | 97.6 | 77.4 | — | — | — | 88.6 | — | |
| DriveFineSensors=1xC, Planner Type=VLA-based Planner, Reinforcement Learning Training=true2026.06 | 98.7 | 97.3 | 88.2 | 97.8 | 98.4 | 99.8 | 98.8 | 97.7 | 84.7 | — | — | — | 87.1 | — | |
| SGDriveSensors=1xC, Planner Type=VLA-based Planner2026.06 | 98.6 | 94.3 | 86 | 97.9 | 98.3 | 99.9 | 99.5 | 96.1 | 85.9 | — | — | — | 86.2 | — | |
| DriveWorld-VLAMethod Category=VLA-based Methods2026.06 | 98.6 | 99.1 | 87.4 | 97.9 | 97.8 | — | 99.6 | 97 | 78.6 | — | — | 99.8 | 86.8 | — | |
| Senna-22026.03 | 98.5 | 97.8 | 88.1 | 97.5 | 98.6 | — | 99.5 | 97 | 88.4 | 86.6 | — | 99.8 | — | — | |
| DriveVLA-W02026.04 | 98.5 | 99.1 | 86.4 | 98.1 | 97.9 | 99.7 | 98 | 93.2 | 58.9 | 86.1 | — | — | — | — | |
| WoTEPerception=true2026.05 | 98.5 | 96.8 | 86.1 | 97.9 | 98.3 | 99.8 | 98.8 | 95.5 | 82.9 | — | — | — | 87.7 | — | |
| EponaV2Perception=false2026.05 | 98.5 | 97.4 | 87.9 | 98.1 | 98.2 | 99.9 | 99.5 | 97.7 | 77.4 | — | — | — | 88.9 | — | |
| DiffusionDriveData Strategy=SimScale-planner, Real Samples=85K, Synthetic Samples=237K2026.05 | 98.5 | 97.1 | 87.4 | 97.8 | 98.3 | 99.8 | 99.6 | 98 | 87.5 | — | — | — | 85.9 | — | |
| DiffusionDriveData Strategy=AutoScale, Real Samples=85K, Synthetic Samples=50K2026.05 | 98.5 | 96.8 | 87.4 | 97.7 | 98.3 | 99.8 | 99.6 | 97.6 | 87.7 | — | — | — | 85.6 | — | |
| DiffusionDriveData Strategy=AutoScale, Real Samples=85K, Synthetic Samples=100K2026.05 | 98.5 | 97 | 87.5 | 97.7 | 98.3 | 99.8 | 99.6 | 97.4 | 87.9 | — | — | — | 85.9 | — | |
| DriveVLA-W0Sensors=1xC, Planner Type=WAM-based Planner, Training on full navtrain split=true2026.06 | 98.5 | 99.1 | 86.4 | 98.1 | 97.9 | 99.7 | 98 | 93.2 | 58.9 | — | — | — | 86.1 | — | |
| MetisSensors=1xC, Planner Type=WAM-based Planner, Best-of-N strategy=N=62026.06 | 98.5 | 97.5 | 87.9 | 97.8 | 98.4 | 99.8 | 99.6 | 98 | 90 | — | — | — | 90.3 | — | |
| DriveVLA-W0Method Category=VLA-based Methods2026.06 | 98.5 | 99.1 | 86.4 | 98.1 | 97.9 | — | 98 | 93.2 | 58.9 | — | — | 99.7 | 86.1 | — | |
| MindDriveBackbone=ResNet-342025.12 | 98.4 | 95.6 | 86.7 | 97.5 | 100 | 99.8 | 99.3 | 96.5 | 96.8 | 84.2 | 88.9 | — | — | — | |
| DriveVLA-W0Perception=false2026.05 | 98.4 | 95.2 | 86.6 | 97.9 | 98.3 | 99.9 | 99.4 | 97.8 | 82.7 | — | — | — | 86.9 | — | |
| Hydra-MDP++2026.05 | 98.4 | 98 | 87.5 | 97.7 | 98.3 | — | 99.4 | 95.3 | 77.4 | — | — | 99.8 | — | 85.1 | |
| DiffusionDriveData Strategy=Real Data, Real Samples=85K2026.05 | 98.4 | 95.5 | 87.5 | 97.5 | 98.4 | 99.8 | 99.5 | 96.9 | 87.7 | — | — | — | 84.2 | — | |
| DiffusionDriveData Strategy=SimScale-recovery, Real Samples=85K, Synthetic Samples=147K2026.05 | 98.4 | 96.7 | 87.6 | 97.5 | 98.3 | 99.8 | 99.6 | 97.5 | 87.1 | — | — | — | 85.4 | — | |
| MetisSensors=1xC, Planner Type=WAM-based Planner2026.06 | 98.4 | 97.2 | 87.8 | 97.7 | 98.4 | 99.8 | 99.6 | 97.8 | 88 | — | — | — | 89.5 | — | |
| GraphWorldMethod Category=World-Model-based Methods2026.06 | 98.4 | 98.8 | 85.9 | 97.9 | 97.8 | — | 99.1 | 96 | 74.6 | — | — | 99.1 | 89.5 | — | |
| ARTEMISBackbone=ResNet-342025.12 | 98.3 | 95.1 | 81.5 | 97.4 | 100 | 99.8 | 98.6 | 96.5 | 98.3 | 83.1 | 87 | — | — | — | |
| ARTEMIS2026.03 | 98.3 | 95.1 | 81.5 | 97.4 | 98.3 | — | 98.6 | 96.5 | — | 83.1 | — | 99.8 | — | — | |
| ARTEMIS2026.04 | 98.3 | 95.1 | 81.5 | 97.4 | 98.3 | 99.8 | 98.6 | 96.5 | — | 83.1 | — | — | — | — | |
| ARTEMIS2026.05 | 98.3 | 95.1 | 81.5 | 97.4 | 98.3 | — | 98.6 | 96.5 | 98.3 | — | — | 99.8 | — | 83.1 | |
| TransFuserData Strategy=SimScale-planner, Real Samples=85K, Synthetic Samples=237K2026.05 | 98.3 | 95.6 | 87.1 | 97.5 | 98.3 | 99.8 | 99.6 | 97.2 | 88.2 | — | — | — | 84.4 | — | |
| ARTEMISSensors=3xC+L, Planner Type=Traditional E2E-based Planner2026.06 | 98.3 | 95.1 | 81.5 | 97.4 | 98.3 | 99.8 | 98.6 | 96.5 | — | — | — | — | 83.1 | — | |
| ReCogDriveSensors=1xC, Planner Type=VLA-based Planner, Reinforcement Learning Training=true2026.06 | 98.3 | 95.2 | 87.1 | 97.5 | 98.3 | 99.8 | 99.5 | 96.6 | 86.5 | — | — | — | 83.6 | — | |
| RecogdriveMethod Category=VLA-based Methods2026.06 | 98.3 | 95.2 | 87.1 | 97.5 | 99.5 | — | 98.3 | 96.6 | 86.5 | — | — | 99.8 | 83.6 | — | |
| DiffusionDrive2026.03 | 98.2 | 95.9 | 87.5 | 97.3 | 98.3 | — | 99.4 | 96.8 | 87.7 | 84.5 | — | 99.8 | — | — | |
| DiffusionDrive2026.04 | 98.2 | 95.9 | 87.5 | 97.3 | 98.3 | 99.8 | 99.4 | 96.8 | 87.7 | 84.5 | — | — | — | — | |
| DiffusionDrivePerception=true2026.05 | 98.2 | 96.2 | 87.4 | 97.3 | 98.4 | 99.8 | 99.5 | 96.9 | 87.7 | — | — | — | 88.2 | — | |
| DiffusionDriveSensors=3xC+L, Planner Type=Traditional E2E-based Planner2026.06 | 98.2 | 95.9 | 87.5 | 97.3 | 98.3 | 99.8 | 99.4 | 96.8 | 87.7 | — | — | — | 84.5 | — | |
| DiffusionDriveMethod Category=E2E-based Methods2026.06 | 98.2 | 95.9 | 87.5 | 97.3 | 98.3 | — | 99.4 | 96.8 | 87.7 | — | — | 99.8 | 84.5 | — | |
| SparseDriveV22026.05 | 98.1 | 98.1 | 91.1 | 97.3 | 98.2 | — | 99.6 | 96.9 | 78.4 | — | — | 99.8 | 90.1 | 86.7 | |
| TransFuserData Strategy=AutoScale, Real Samples=85K, Synthetic Samples=50K2026.05 | 98.1 | 95.8 | 86.7 | 97.5 | 98.3 | 99.9 | 99.4 | 97.4 | 87.3 | — | — | — | 84.1 | — | |
| Latent-WAMMethod Category=World-Model-based Methods2026.06 | 98.1 | 97.3 | 87.7 | 97.3 | 98.1 | — | 99.6 | 97.6 | 87.3 | — | — | 99.8 | 89.3 | — | |
| Hydra-MDP++Backbone=ResNet-342025.12 | 97.9 | 96.5 | 79.2 | 93.4 | 100 | 100 | 98.9 | 67.2 | 97.7 | 80.6 | 86.6 | — | — | — | |
| ResAD2026.03 | 97.8 | 97.2 | 88.2 | 96.9 | 98.4 | — | 99.5 | 97 | 88.2 | 85.5 | — | 99.8 | — | — | |
| DriveSuprim2026.05 | 97.8 | 97.9 | 90.6 | 97.1 | 98.3 | — | 99.5 | 96.6 | 77.9 | — | — | 99.9 | — | 86 | |
| TransFuserData Strategy=AutoScale, Real Samples=85K, Synthetic Samples=100K2026.05 | 97.8 | 96.1 | 88.3 | 97 | 98.3 | 99.8 | 99.7 | 97 | 87.6 | — | — | — | 84.5 | — | |
| World4DriveSensors=3xC, Planner Type=Traditional E2E-based Planner2026.06 | 97.8 | 96.3 | 88.3 | 97.1 | 98 | 99.8 | 99.4 | 97.7 | 53.9 | — | — | — | 84.8 | — | |
| WorldRFTSensors=3xC, Planner Type=Traditional E2E-based Planner, Reinforcement Learning Training=true2026.06 | 97.8 | 96.5 | 88.5 | 97 | 98.1 | 99.8 | 99.5 | 97.4 | 69.1 | — | — | — | 86.7 | — | |
| TransfuserBackbone=ResNet-342025.12 | 97.7 | 92.8 | 79.2 | 92.8 | 100 | 99.9 | 98.3 | 67.6 | 95.3 | 77.8 | 84 | — | — | — | |
| DiffusionDriveV22026.05 | 97.7 | 96.6 | 88.9 | 97.2 | 97.8 | — | 99.2 | 96 | 91 | — | — | 99.8 | 87.5 | 85.5 | |
| TransFuserData Strategy=Real Data, Real Samples=85K2026.05 | 97.7 | 94 | 87.2 | 96.7 | 98.3 | 99.8 | 99.3 | 95.5 | 82.9 | — | — | — | 81.5 | — | |
| DiffusionDriveV2Method Category=E2E-based Methods2026.06 | 97.7 | 96.6 | 88.9 | 97.2 | 97.8 | — | 99.2 | 96 | 91 | — | — | 99.8 | 87.5 | — | |
| GTRS-DenseSensors=3xC, Planner Type=Traditional E2E-based Planner2026.06 | 97.6 | 97.5 | 87.9 | 97 | 97.5 | 99.9 | 99 | 95.9 | 55.9 | — | — | — | 82.3 | — | |
| Hydra-MDPBackbone=ResNet-342025.12 | 97.5 | 96.3 | 80.1 | 93 | 100 | 99.9 | 98.3 | 65.5 | 97.4 | 79.8 | 86.5 | — | — | — | |
| DriveSuprim2026.03 | 97.5 | 96.5 | 88.4 | 96.6 | 98.3 | — | 99.4 | 95.5 | 77 | 83.1 | — | 99.6 | — | — | |
| Drivesuprim2026.04 | 97.5 | 96.5 | 88.4 | 96.6 | 98.3 | 99.6 | 99.4 | 95.5 | 77 | 83.1 | — | — | — | — | |
| DriveSuprimSensors=3xC, Planner Type=Traditional E2E-based Planner2026.06 | 97.5 | 96.5 | 88.4 | 96.6 | 98.3 | 99.6 | 99.4 | 95.5 | 77 | — | — | — | 83.1 | — | |
| DriveSuprimMethod Category=E2E-based Methods2026.06 | 97.5 | 96.5 | 88.4 | 96.6 | 98.3 | — | 99.4 | 95.5 | 77 | — | — | 99.6 | 83.1 | — | |
| VADv2Backbone=ResNet-342025.12 | 97.3 | 91.7 | 77.6 | 92.7 | 100 | 99.9 | 98.2 | 66 | 97.4 | 76.6 | 83 | — | — | — | |
| Hydra-MDP++2026.03 | 97.2 | 97.5 | 83.1 | 96.5 | 98.2 | — | 99.4 | 94.4 | 70.9 | 81.4 | — | 99.6 | — | — | |
| Hydra-MDP++2026.04 | 97.2 | 97.5 | 83.1 | 96.5 | 98.2 | 99.6 | 99.4 | 94.4 | 70.9 | 81.4 | — | — | — | — | |
| TransFuserData Strategy=SimScale-recovery, Real Samples=85K, Synthetic Samples=147K2026.05 | 97.2 | 92.8 | 88.8 | 96.4 | 98.3 | 99.8 | 99.6 | 96.7 | 86.8 | — | — | — | 83.6 | — | |
| Hydra-MDP++Sensors=3xC+L, Planner Type=Traditional E2E-based Planner2026.06 | 97.2 | 97.5 | 83.1 | 96.5 | 98.2 | 99.6 | 99.4 | 94.4 | 70.9 | — | — | — | 81.4 | — | |
| Hydra-MDP++Method Category=E2E-based Methods2026.06 | 97.2 | 97.5 | 83.1 | 96.5 | 98.2 | — | 99.4 | 94.4 | 70.9 | — | — | 99.6 | 81.4 | — | |
| EponaSensors=1xC, Planner Type=WAM-based Planner2026.06 | 97.1 | 95.7 | 88.6 | 96.3 | 98 | 99.7 | 99.3 | 97 | 67.8 | — | — | — | 85.1 | — | |
| Transfuser2026.03 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | — | 97.8 | 92.7 | 87.2 | 76.7 | — | 99.7 | — | — | |
| TransFuser2026.04 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | 99.7 | 97.8 | 92.7 | 87.2 | 76.7 | — | — | — | — | |
| TranfuserPerception=true2026.05 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | 99.7 | 97.8 | 92.7 | 87.2 | — | — | — | 84 | — | |
| Transfuser2026.05 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | — | 97.8 | 92.7 | 87.2 | — | — | 99.7 | — | 76.7 | |
| TransFuserSensors=3xC+L, Planner Type=Traditional E2E-based Planner2026.06 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | 99.7 | 97.8 | 92.7 | 87.2 | — | — | — | 76.7 | — | |
| TransFuserMethod Category=E2E-based Methods2026.06 | 96.9 | 89.9 | 87.1 | 95.4 | 98.3 | — | 97.8 | 92.7 | 87.2 | — | — | 99.7 | 76.7 | — | |
| SpanVLAEvaluation protocol=One-shot2026.04 | 96.4 | 89.4 | 87.4 | 95.8 | 98.2 | 99.8 | 97.8 | 94.5 | 81.6 | 79.4 | — | — | — | — | |
| Ego-MLP2026.05 | 93.1 | 77.9 | 86 | 91.5 | 98.3 | — | 92.7 | 89.4 | 85.4 | — | — | 99.6 | — | 64 |