Play2Perfect checkpoints
State-based policy checkpoints for Play2Perfect (arXiv: 2606.26428).
play/model.pthβ Stage-1 "play" pretrained policy.tight_insertion/model.pthβ Stage-2: tight insertion (L-peg, 0.5 mm).beam_assembly_step1/model.pthβ Stage-2: beam assembly step 1.beam_assembly_step2/model.pthβ Stage-2: beam assembly step 2.screwing/model.pthβ Stage-2: screwing (furniture leg).fork_insertion/model.pthβ Stage-2: YCB fork into a flatware holder (handle first, 1 mm clearance).plug_insertion/model.pthβ Stage-2: USB-C wall plug into a power-strip socket (0.5 mm clearance).
The checkpoints above are for the KUKA iiwa 14 + Sharpa hand
(task Isaacsimenvs-PreciseAssembly-Direct-v0).
Flexiv Rizon 4s + right Wuji Hand 2
Under flexiv_wuji_right_human_like/
(task Isaacsimenvs-PreciseAssemblyFlexivWujiRight-Direct-v0):
play/model.pthβ Stage-1 "play" pretrained policy.tight_insertion/model.pth,beam_assembly_step1/model.pth,beam_assembly_step2/model.pth,fork_insertion/model.pth,plug_insertion/model.pthβ Stage-2 policies finetuned from it (domain randomization off).
There is no Flexiv screwing policy yet. These are snapshots of runs that are still training and may be updated.
Download
From the Play2Perfect repo:
python download_checkpoints.py # KUKA + Sharpa
python download_checkpoints.py --robot flexiv_wuji_right_human_like # Flexiv + Wuji