Closed-loop Planning on nuPlan 14 (test)
95.02NRFlowDrive*
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FlowDrive*Type=Rule-based & Hybrid2025.09 | 95.02 | 93.96 | — | |
| Pluto-SoEType=Hybrid, Time (ms)=2592025.12 | 94.82 | 90.47 | — | |
| DiffusionType=Hybrid2025.12 | 94.8 | 91.75 | — | |
| Diffusion PlannerType=Rule-based & Hybrid, Refinement=Enabled2025.09 | 94.8 | 91.75 | — | |
| Diffusion Planner w/ refineType=Rule-based & Hybrid2026.02 | 94.8 | 91.75 | — | |
| Log-ReplayType=Rule2025.12 | 94.03 | 75.86 | — | |
| Log-replayType=Expert2025.09 | 94.03 | 75.86 | — | |
| Log-replayType=Expert2026.02 | 94.03 | 75.86 | — | |
| Log-replay (Expert)2026.07 | 94.03 | 75.86 | — | |
| PlutoType=Hybrid, Time (ms)=2592025.12 | 92.55 | 90.29 | — | |
| PLUTOType=Rule-based & Hybrid2025.09 | 92.23 | 90.29 | — | |
| PLUTOType=Rule-based & Hybrid2026.02 | 92.23 | 90.29 | — | |
| PLUTO2026.07 | 92.23 | 90.29 | — | |
| Pluto w/o-SoE1Type=Learning, Time (ms)=1072025.12 | 91.39 | 82.29 | — | |
| FlowDriveType=Learning-based2025.09 | 91.02 | 87.28 | — | |
| Pluto w/o-SoE2Type=Learning, Time (ms)=1072025.12 | 90.85 | 83.07 | — | |
| Pluto w/oType=Learning, Time (ms)=1072025.12 | 90.67 | 78.06 | — | |
| FlowDrive-Type=Learning-based2025.09 | 90.19 | 83.47 | — | |
| PDM-HybridType=Rule-based & Hybrid2025.09 | 90.1 | 91.28 | — | |
| PDM-HybridType=Rule-based & Hybrid2026.02 | 90.1 | 91.28 | — | |
| PDM-Hybrid2026.07 | 90.1 | 91.28 | — | |
| PDM-ClosedType=Rule, Time (ms)=1222025.12 | 90.05 | 91.64 | — | |
| PDM-ClosedType=Rule-based & Hybrid2025.09 | 90.05 | 91.63 | — | |
| PDM-ClosedType=Rule-based & Hybrid2026.02 | 90.05 | 91.63 | — | |
| PDM-Closed2026.07 | 90.05 | 91.63 | — | |
| RAPiDType=Learning-based2026.02 | 89.98 | 81.86 | — | |
| PLUTOType=Learning-based, Refinement=None2025.09 | 89.9 | 78.62 | — | |
| PLUTO w/o refineType=Learning-based, Pre-computed reference paths=true2026.02 | 89.9 | 78.62 | — | |
| Flow Planner2026.07 | 89.88 | 82.93 | — | |
| Diffusion w/oType=Learning, Time (ms)=1592025.12 | 89.4 | 82.67 | — | |
| Diffusion PlannerType=Learning-based2025.09 | 89.19 | 82.93 | — | |
| DiffusionPlannerType=Learning-based2026.02 | 89.19 | 82.93 | — | |
| Diffusion Planner2026.07 | 89.19 | 82.93 | — | |
| planTF-SoEType=Learning, Time (ms)=782025.12 | 88.06 | 78.88 | — | |
| Intent2Drive2026.07 | 87.2 | 81.35 | — | |
| PlanTFType=Learning-based2025.09 | 85.62 | 79.58 | — | |
| PlanTFType=Learning-based2026.02 | 85.62 | 79.58 | — | |
| PlanTF2026.07 | 85.62 | 79.58 | — | |
| planTFType=Learning, Time (ms)=782025.12 | 85.6 | 78.86 | — | |
| GameFormerType=Rule-based & Hybrid2025.09 | 83.88 | 82.05 | — | |
| GameFormerType=Rule-based & Hybrid2026.02 | 83.88 | 82.05 | — | |
| GameFormer2026.07 | 83.88 | 82.05 | — | |
| IDMType=Rule, Time (ms)=412025.12 | 70.39 | 72.42 | — | |
| IDMType=Rule-based & Hybrid2025.09 | 70.39 | 74.42 | — | |
| IDMType=Rule-based & Hybrid2026.02 | 70.39 | 74.42 | — | |
| IDM2026.07 | 70.39 | 74.42 | — | |
| Raster-SoEType=Learning, Time (ms)=372025.12 | 69.89 | 72.86 | — | |
| RasterType=Learning, Time (ms)=372025.12 | 65.87 | 69.93 | — | |
| PDM-OpenType=Learning, Time (ms)=652025.12 | 52.91 | 58.1 | — | |
| PDM-OpenType=Learning-based2025.09 | 52.81 | 57.23 | — | |
| PDM-OpenType=Learning-based, Pre-computed reference paths=true2026.02 | 52.81 | 57.23 | — | |
| PDM-Open-SoEType=Learning, Time (ms)=652025.12 | 52.75 | 59.85 | — | |
| UrbanDriverType=Learning-based2026.02 | 51.83 | 67.15 | — | |
| UrbanDriver2026.07 | 51.83 | 67.15 | — | |
| GameFormerType=Learning-based, Refinement=None2025.09 | 11.36 | 9.31 | — | |
| GameFormer w/o refineType=Learning-based2026.02 | 11.36 | 9.31 | — | |
| Diffusion PlannerType=Rule-based & Hybrid, Refinement=true2025.11 | — | — | 94.13 | |
| Diffusion PlannerType=Learning-based2025.11 | — | — | 88.84 | |
| Diffusion PlannerReactive setting=Non-reactive (NR), Refinement module=with refinement2026.04 | — | — | 94.8 | |
| Diffusion PlannerReactive setting=Reactive (R), Refinement module=with refinement2026.04 | — | — | 91.75 | |
| Diffusion PlannerReactive setting=Non-reactive (NR)2026.04 | — | — | 89.19 | |
| Diffusion PlannerReactive setting=Reactive (R)2026.04 | — | — | 82.93 | |
| GameFormerReactive setting=Non-reactive (NR)2026.04 | — | — | 83.88 | |
| GameFormerReactive setting=Reactive (R)2026.04 | — | — | 82.05 | |
| GC-PGPType=Learning-based2025.11 | — | — | 61.16 | |
| IDMType=Rule-based & Hybrid2025.11 | — | — | 70.39 | |
| IDMReactive setting=Non-reactive (NR)2026.04 | — | — | 70.39 | |
| IDMReactive setting=Reactive (R)2026.04 | — | — | 74.42 | |
| LAPType=Rule-based & Hybrid, Refinement=true, Solver order (o)=1, Denoising steps (s)=12025.11 | — | — | 93.9 | |
| LAPType=Rule-based & Hybrid, Refinement=true, Solver order (o)=1, Denoising steps (s)=22025.11 | — | — | 94.4 | |
| LAPType=Learning-based, Solver order (o)=1, Denoising steps (s)=12025.11 | — | — | 89.85 | |
| LAPType=Learning-based, Solver order (o)=1, Denoising steps (s)=22025.11 | — | — | 90.42 | |
| Log-replayReactive setting=Non-reactive (NR)2026.04 | — | — | 94.03 | |
| Log-replayReactive setting=Reactive (R)2026.04 | — | — | 75.86 | |
| Log-ReplayType=Expert2025.11 | — | — | 94.03 | |
| MISTYReactive setting=Non-reactive (NR), Refinement module=with refinement2026.04 | — | — | 91.24 | |
| MISTYReactive setting=Reactive (R), Refinement module=with refinement2026.04 | — | — | 90.7 | |
| MISTYReactive setting=Non-reactive (NR), Refinement module=without refinement2026.04 | — | — | 78.88 | |
| MISTYReactive setting=Reactive (R), Refinement module=without refinement2026.04 | — | — | 82.25 | |
| PDM-ClosedType=Rule-based & Hybrid2025.11 | — | — | 90.05 | |
| PDM-ClosedReactive setting=Non-reactive (NR)2026.04 | — | — | 90.05 | |
| PDM-ClosedReactive setting=Reactive (R)2026.04 | — | — | 91.63 | |
| PDM-HybridType=Rule-based & Hybrid2025.11 | — | — | 90.1 | |
| PDM-OpenType=Learning-based, Pre-searched reference lines=true2025.11 | — | — | 52.81 | |
| PlanTFReactive setting=Non-reactive (NR)2026.04 | — | — | 85.62 | |
| PlanTFReactive setting=Reactive (R)2026.04 | — | — | 79.58 | |
| PLUTOType=Rule-based & Hybrid2025.11 | — | — | 92.6 | |
| PLUTOType=Learning-based, Refinement=false, Pre-searched reference lines=true2025.11 | — | — | 90.11 | |
| PLUTOReactive setting=Non-reactive (NR)2026.04 | — | — | 92.23 | |
| PLUTOReactive setting=Reactive (R)2026.04 | — | — | 90.29 | |
| PLUTOReactive setting=Non-reactive (NR), Refinement module=without refinement2026.04 | — | — | 89.9 | |
| PLUTOReactive setting=Reactive (R), Refinement module=without refinement2026.04 | — | — | 78.62 | |
| UrbanDriverType=Learning-based2025.11 | — | — | 51.83 |