Datasetpublic

lance-format/agibotworld-beta-rgb-lance

AgiBotWorld-Beta (LeRobot lance format) agibot-world/AgiBotWorld-Beta converted to the LeRobot lance storage format, uploaded in coordination with the AgiBot team. 160,454 episodes, 286,556,463 frames, 8 RGB cameras (AV1, copied from the source without re-encoding), state[20] and action[22] at 30 fps. Read it in place, no download needed (lerobot with the lancedb extra): from lerobot.datasets import LeRobotDataset ds = LeRobotDataset("lance-format/agibotworld-beta-rgb-lance")… See the full description on the dataset page: https://huggingface.co/datasets/lance-format/agibotworld-beta-rgb-lance.

sourceHugging Facecc-by-nc-sa-4.0updated 20d agoView on Hugging Face

0likes1.7kdownloads

Dataset Card

AgiBotWorld-Beta (LeRobot lance format)

agibot-world/AgiBotWorld-Beta converted to the LeRobot lance storage format, uploaded in coordination with the AgiBot team. 160,454 episodes, 286,556,463 frames, 8 RGB cameras (AV1, copied from the source without re-encoding), state[20] and action[22] at 30 fps.

Read it in place, no download needed (lerobot with the lancedb extra):

python

from lerobot.datasets import LeRobotDataset

ds = LeRobotDataset("lance-format/agibotworld-beta-rgb-lance")
item = ds[0]  # 8 decoded cameras + state/action/task, fetched on demand

Works directly with lerobot-train, eval_split, and lerobot-dataset-viz (rerun or foxglove). Note: sustained multi-worker training against Hub storage can hit rate limits; for heavy training, mirror to your own object store and pass it as root=.

License: CC BY-NC-SA 4.0, same as the source. Credit AgiBot / OpenDriveLab.