Written for Ugur Oezdemir at Bonsai Robotics. The dataset read here is apairo-robotics/tartandrive_offroad. It is not Bonsai Robotics's data, and it is on this page because TartanDrive, the public off-road driving corpus, the readable equivalent of the terrain Bonsai operates on.
55 episodes and 123,786 frames at 10 fps. One task string, drive off-road, covers all 55 episodes, so nothing in the release distinguishes a gentle track from a hard one. Episode length runs from 18 frames to 8,713, a range of nearly five hundred to one, which means episode count says very little about what the corpus holds.
Read from apairo-robotics/tartandrive_offroad. Where a figure came from the release metadata rather than the frames, the mode line below says so.
| property | value |
|---|---|
| episodes | 55 |
| frames | 123,786 |
| frame rate | 10 fps |
| robot | tartandrive |
| action dimensions | 2 |
| cameras | 1 |
| distinct task strings | 1 |
| episode length, shortest | 18 frames |
| episode length, median | 1,831 frames |
| episode length, longest | 8,713 frames |
Measurement mode, meta only, frame data is 4.1 GB and was not downloaded.
A trailing hold is the number of frames at the end of an episode during which the commanded action does not change by more than 1e-4 on any component. Seconds come from the frame rate each release declares. Every corpus below was read the same way by the same code.
| public corpus | how it was produced | episodes | any trailing hold | ten frames or more | longest |
|---|---|---|---|---|---|
| lerobot/aloha_sim_insertion_scripted | scripted sim | 50 | 100.0% | 86.0% | 21 frames , 0.42s |
| RajatDandekar/so101_bimanual_napkin_fold | real bimanual teleop | 65 | 95.4% | 75.4% | 130 frames , 4.33s |
| weblucas/sponge_ball_2_yellow_tray_lerobot | real single arm teleop | 30 | 26.7% | 6.7% | 21 frames , 0.70s |
| lerobot/aloha_sim_insertion_human | sim human teleop | 50 | 6.0% | 0.0% | 1 frames , 0.02s |
| lerobot/ucsd_pick_and_place_dataset | sim, fixed window | 1,355 | 0.1% | 0.0% | 2 frames , 0.40s |
| RajatDandekar/eval_so101_pi0_bimanual_napkin_fold | policy rollouts | 6 | 0.0% | 0.0% | 0 frames , 0.00s |
The ordering is the finding. A script stops commanding when it is done and the last rows repeat. A person keeps making small corrections, so a short take ends clean while a long bimanual take ends with the operator holding the fold in place. A policy rollout ends the moment the episode is cut. If somebody quotes you a single defect rate for idle frames across a mixed corpus, that number is averaging four different things.
Each corpus is a meta directory plus one or more parquet files. Read data/chunk-000/file-000.parquet, group rows by the episode ranges in meta/episodes, and count trailing rows where the action vector does not change. No credentials and no download beyond the frame data itself.
Thirty eight public robot datasets, one at a time. Each report measures one property across every episode, removes the part that scores worst, and retrains on what is left. The corpus figures and the cuts are computed from the releases. The lift column on the results page is read from each report's fine tune table rather than from a training run, and that page says so.
The one result measured on physical hardware is Stera-10M. Two arms on a four arm SO-ARM101 bench, identical per task fine tunes, 127 unpaired rollouts per cell, and the only difference between the arms is whether the human video pre-training stage happened.
Everything above is a public release. The same four steps run on your data, and the first three cost nothing.
A few hundred episodes under whatever terms you want is enough for the first three.