Trajectory planning on NAVSIM v2 (navhard)
98.9NC RateGTRS-Aug
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| GTRS-AugImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 12025.12 | 98.9 | 95.1 | 99.2 | 99.6 | 76.1 | 99.1 | 94.7 | 97.6 | 54.2 | — | |
| TrajMoE-MoEImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 12025.12 | 98.8 | 96.6 | 100 | 99.7 | 75 | 98.6 | 95.5 | 97.5 | 50.6 | — | |
| GTRS-DenseImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 12025.12 | 98.7 | 95.8 | 99.4 | 99.3 | 72.8 | 98.7 | 95.1 | 96.9 | 40.4 | — | |
| GTRS-AugImg. Res.=256 x 1024, Backbone=DINOv3-L, Stage=Stage 12025.12 | 98.3 | 97.1 | 99.3 | 99.7 | 77.4 | 98.6 | 96 | 97.5 | 54.2 | — | |
| TrajMoE-GRPOImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 12025.12 | 98 | 96.6 | 99.2 | 99.3 | 77.3 | 98 | 96 | 97.5 | 54.2 | — | |
| GTRS-AugImg. Res.=256 x 1024, Backbone=DINOv3-L-M, Stage=Stage 12025.12 | 97.6 | 96.4 | 99.2 | 99.6 | 76 | 98.7 | 96 | 97.3 | 54.7 | — | |
| WoTESensor modality=multimodal (camera + LiDAR)2025.12 | 97.4 | 88.2 | 97.8 | 99.3 | 82.7 | 96.4 | 90.9 | 97.3 | 68 | 66.7 | |
| WoTEBackbone=ResNet34, Stage=Stage 12026.05 | 97.4 | 88.2 | 97.7 | 99.3 | 82.7 | 96.4 | 90.8 | 97.3 | 68 | — | |
| DriveLaWPerception=false, Stage=12026.05 | 97.3 | 89.1 | 99.2 | 99.6 | 84.3 | 97.1 | 96.2 | 97.8 | 67.6 | 30.6 | |
| EponaV2Perception=false, Stage=12026.05 | 97.3 | 90.7 | 99.4 | 100 | 83.3 | 97.3 | 97.3 | 97.6 | 60.9 | 36.1 | |
| IDOLBackbone=ResNet34, Stage=Stage 12026.05 | 97.2 | 89.6 | 98 | 99.6 | 82.3 | 96.9 | 95.3 | 97.6 | 72.9 | — | |
| RAPPerception=true, Stage=12026.05 | 97.1 | 94.4 | 98.8 | 99.8 | 83.9 | 96.9 | 94.7 | 96.4 | 66.2 | 39.6 | |
| MindDriveSensor modality=multimodal (camera + LiDAR)2025.12 | 97 | 89.7 | 98.6 | 99.3 | 83.7 | 95.8 | 96 | 97.5 | 71.1 | 72.7 | |
| DriveVLA-W0Perception=false, Stage=12026.05 | 96.8 | 83.3 | 99 | 99.6 | 84.6 | 95.3 | 96.4 | 97.6 | 78.2 | 24.4 | |
| LTFv6Perception=true, Stage=12026.05 | 96.6 | 86.7 | 99.2 | 99.6 | 84.5 | 95.1 | 94.4 | 97.8 | 76.4 | 31.9 | |
| GuideFlowBackbone=ResNet34, Stage=Stage 12026.05 | 96.6 | 80.5 | 96.3 | 99.3 | 82.3 | 94.9 | 91.5 | 97.7 | 67.8 | — | |
| TransfuserSensor modality=multimodal (camera + LiDAR)2025.12 | 96.3 | 74.6 | 98.4 | 99.3 | 82.9 | 93.7 | 92.7 | 97.5 | 78.2 | 60.5 | |
| LTFWeight source=reproduced from official weights2025.12 | 96.2 | 79.5 | 99.1 | 99.5 | 84.1 | 95.1 | 94.2 | 97.5 | 79.1 | 62.3 | |
| LTFPerception=true, Stage=12026.05 | 96.2 | 79.6 | 99.1 | 99.6 | 84.1 | 95.1 | 94.2 | 97.6 | 79.1 | 25.1 | |
| LTFBackbone=ResNet34, Stage=Stage 12026.05 | 96.2 | 79.5 | 99.1 | 99.5 | 84.1 | 95.1 | 94.2 | 97.5 | 79.1 | — | |
| DiffusionDriveBackbone=ResNet34, Stage=Stage 12026.05 | 96 | 79.7 | 97.4 | 99.5 | 81.3 | 93.1 | 90.8 | 96.8 | 73.8 | — | |
| DiffusionDriveSensor modality=multimodal (camera + LiDAR)2025.12 | 95.9 | 84 | 98.6 | 99.8 | 84.4 | 96 | 95.1 | 97.6 | 71.1 | 63.2 | |
| GTRS-DPBackbone=ResNet34, Stage=Stage 12026.05 | 94.7 | 78.8 | 96.1 | 99.5 | 83 | 94.4 | 92 | 97.5 | 72.8 | — | |
| PDM-ClosedBackbone=-, Stage=Stage 12026.05 | 94.4 | 98.8 | 100 | 99.5 | 100 | 93.5 | 99.3 | 87.7 | 36 | — | |
| GTRS-DenseImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 22025.12 | 91.4 | 89.2 | 94.4 | 98.8 | 69.5 | 90.1 | 54.6 | 94.1 | 49.7 | 41.7 | |
| TrajMoE-MoEImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 22025.12 | 90 | 88 | 91.3 | 98.8 | 79.2 | 87.6 | 53.6 | 96.5 | 56.8 | 43.8 | |
| GTRS-AugImg. Res.=256 x 1024, Backbone=DINOv3-L-M, Stage=Stage 22025.12 | 89 | 88.3 | 91.7 | 98.9 | 80.9 | 87.2 | 55.6 | 96.8 | 55.3 | 43.1 | |
| PDM-ClosedBackbone=-, Stage=Stage 22026.05 | 88.1 | 90.6 | 96.3 | 98.5 | 100 | 83.1 | 73.7 | 91.5 | 25.4 | 51.3 | |
| GTRS-AugImg. Res.=256 x 1024, Backbone=DINOv3-L, Stage=Stage 22025.12 | 88 | 88.9 | 90.6 | 98.1 | 84.9 | 84 | 55.4 | 96.9 | 52.5 | 43.8 | |
| TrajMoE-GRPOImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 22025.12 | 88 | 87.9 | 90.8 | 99 | 82.5 | 85.3 | 52.4 | 97.1 | 51.2 | 42.7 | |
| GTRS-AugImg. Res.=512 x 2048, Backbone=V2-99, Stage=Stage 22025.12 | 87.9 | 88.8 | 89.6 | 98.8 | 80.3 | 86 | 53.5 | 97.1 | 56.1 | 42.1 | |
| GuideFlowBackbone=ResNet34, Stage=Stage 22026.05 | 87.3 | 76.7 | 88.8 | 99.2 | 84.3 | 85.1 | 49.7 | 93.1 | 44.5 | 27.1 | |
| IDOLBackbone=ResNet34, Stage=Stage 22026.05 | 84.9 | 81.7 | 87.7 | 98.5 | 84.7 | 81.1 | 50 | 95.4 | 64 | 38 | |
| EponaV2Perception=false, Stage=22026.05 | 83.6 | 78 | 88 | 98.9 | 86 | 80.3 | 50.1 | 96.1 | 52 | — | |
| RAPPerception=true, Stage=22026.05 | 83.2 | 83.9 | 87.4 | 98 | 86.9 | 80.4 | 52.3 | 95.2 | 52.4 | — | |
| DriveLaWPerception=false, Stage=22026.05 | 82.5 | 67.6 | 83.5 | 98.1 | 84.8 | 78.5 | 45.8 | 96.4 | 57.3 | — | |
| DiffusionDriveBackbone=ResNet34, Stage=Stage 22026.05 | 82.1 | 72.2 | 88.5 | 98.7 | 85.1 | 78.8 | 49.2 | 89.3 | 71.2 | 24.2 | |
| WoTEBackbone=ResNet34, Stage=Stage 22026.05 | 81.2 | 77.7 | 84.8 | 98.1 | 85.9 | 78.5 | 46.2 | 96.6 | 63.3 | 27.9 | |
| GTRS-DPBackbone=ResNet34, Stage=Stage 22026.05 | 80.3 | 74.4 | 84.9 | 98 | 81.9 | 78.8 | 45.4 | 96.7 | 70.1 | 23.8 | |
| LTFv6Perception=true, Stage=22026.05 | 79.9 | 75.6 | 86.3 | 97.9 | 89.6 | 76.1 | 50.1 | 95.2 | 66.7 | — | |
| LTFPerception=true, Stage=22026.05 | 77.8 | 70.2 | 84.3 | 98.1 | 85.1 | 75.7 | 45.4 | 95.7 | 76 | — | |
| LTFBackbone=ResNet34, Stage=Stage 22026.05 | 77.7 | 70.2 | 84.2 | 98 | 85.1 | 75.6 | 45.4 | 95.7 | 75.9 | 23.1 | |
| DriveVLA-W0Perception=false, Stage=22026.05 | 76.8 | 64.3 | 79.9 | 98.3 | 89.2 | 75 | 46.8 | 95.8 | 53.1 | — |