Does Diffusion Policy run on SO-101?

Capabilities

Caveats

• Community-reported only; not documented upstream or maintainer-tested.
• LeRobot's Diffusion Policy defaults target simulation; real SO-101 deployment in #2506 only worked after changing crop_shape (None or matched to image size), vision backbone and inference steps.

Evidence 3 sources; the cell shows the highest tier

1. T3 · community reportedhub-checkpointsource 2025-08-27verified 2026-10-24 (1d ago)

   config.json type=diffusion; train_config.json dataset.repo_id=ymatari/cleanup-table-2, whose meta/info.json robot_type=so101_follower (20 episodes).

   https://huggingface.co/ymatari/diffusion_so101_cleanup_table_2
2. T3 · community reportedhub-checkpointsource 2026-01-09verified 2026-10-24 (1d ago)

   config.json type=diffusion; train_config.json dataset.repo_id=jliu6718/diffusion-so101-place_brick, whose meta/info.json robot_type=so101_follower (30 episodes).

   https://huggingface.co/jliu6718/diffusion-so101-place_brick
3. T3 · community reportedgithub-issuesource 2025-11-23verified 2026-10-24 (1d ago)

   Issue #2506: a user deployed Diffusion Policy on the SO101 follower after changing the sim-oriented defaults (crop_shape, vision backbone, inference steps); maintainer pkooij confirmed and closed it.

   “I have successfully deployed Diffusion Policy on the SO101 Follower using the 100+ episode dataset, following your guidance.”

   https://github.com/huggingface/lerobot/issues/2506