List your company
ResearchFormerly WizWand

Closed-loop planning on nuPlan 14-hard 1.0 (test)

94.03Overall ScoreLog-replay
3.196426.778250.3673.9418Jun 9, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Log-replayType=Expert, Evaluation Protocol=NR2026.0694.03
DFP-FMType=Learning-based, Evaluation Protocol=NR2026.0679.43
DFPType=Learning-based, Evaluation Protocol=NR2026.0676.91
CoPlannerType=Learning-based, Evaluation Protocol=NR2026.0676.82
Diffusion PlannerType=Learning-based, Evaluation Protocol=NR2026.0675.99
Log-replayType=Expert, Evaluation Protocol=R2026.0675.86
Diffusion Planner*Type=Learning-based, Evaluation Protocol=NR, Note=best reimplementation result2026.0674.26
PLUTOType=Learning-based, Evaluation Protocol=NR2026.0670.03
PlanTFType=Learning-based, Evaluation Protocol=NR2026.0669.7
Diffusion PlannerType=Learning-based, Evaluation Protocol=R2026.0669.22
DFP-FMType=Learning-based, Evaluation Protocol=R2026.0667.94
CoPlannerType=Learning-based, Evaluation Protocol=R2026.0664.47
DFPType=Learning-based, Evaluation Protocol=R2026.0663.56
PlanTFType=Learning-based, Evaluation Protocol=R2026.0661.61
Diffusion Planner*Type=Learning-based, Evaluation Protocol=R, Note=best reimplementation result2026.0661.25
PLUTOType=Learning-based, Evaluation Protocol=R2026.0659.74
UrbanDriverType=Learning-based, Evaluation Protocol=NR2026.0650.4
UrbanDriverType=Learning-based, Evaluation Protocol=R2026.0649.95
PDM-OpenType=Learning-based, Evaluation Protocol=R2026.0635.83
PDM-OpenType=Learning-based, Evaluation Protocol=NR2026.0633.51
GameFormerType=Learning-based, Evaluation Protocol=NR2026.067.08
GameFormerType=Learning-based, Evaluation Protocol=R2026.066.69