List your company
ResearchFormerly WizWand

Robot Manipulation on SimplerEnv RT-1 Visual Matching

86.7Success RateMindWPI
57.68465.21772.7580.283Jul 2, 2026
Updated 4mo ago

Evaluation results

MethodLinks
MindWPISize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0786.7
MindLWPISize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0784.4
FPC-VLASize=7B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0778
MemoryVLASize=7B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0777.7
MindPI w/o PTSize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0775.7
MindWPI w/o PTSize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0775.2
SpatialVLASize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0775.1
MindLPISize=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0774.6
MindPI (Frozen VLM)Size=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0772.7
DD-VLASize=7B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0771.2
MindPI (Full PT)Size=4B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0768.2
OpenVLA-OFTSize=7B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0763
π0+FASTSize=3B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0761.9
π0Size=3B, Fine-tuning Protocol=RT-1-only fine-tuning2026.0758.8