✓ Source card reviewedMITVersion: LeRobot v2
ALOHA Sim Insertion — Human Demonstrations
Human-operated bimanual insertion demonstrations from the ALOHA benchmark, containing 50 episodes and 25,000 frames with dual-arm state and action vectors at 50 Hz.
TZea
Published by Tony Zhao et al.LeRobot / Stanford · Source updated 08 Jun 2026 · Reviewed 17 Jul 2026View dataset files ↗91.3 MB · LeRobot v2Downloads and licence terms are handled by the original publisher on Hugging Face. START HERE
This dataset, explained simply
Two simulated robot arms coordinate to insert an object into a matching fixture.
WHAT CAN I TRAIN?✓ Bimanual imitation learning✓ ACT-style policies✓ Coordination experiments
WHAT DO I NEED?• Python basics• About 100 MB storage• No physical robot required
SUITABILITY✓ Contains robot actions✓ No physical robot required• Check source for language labels
New to the terminology?Episode = one complete attemptObservation = what the robot sees and sensesAction = the command sent to the robot
VERIFICATION RECORD
What EmbodiedData checked
Source statusAvailable at publisherLast checked17 July 2026Dataset versionLeRobot v2Training instructionsNot tested by EmbodiedData
Verification confirms the linked metadata at the stated date. It is not a guarantee that upstream files, dependencies or licences will remain unchanged.
Dataset previewIllustrative image—not an original dataset frame
Preview onlyEXAMPLE OF THE TASKInsert the peg using both grippers while keeping the fixture stable.
DATASET OVERVIEW
What the publisher reports
Human-operated bimanual insertion demonstrations from the ALOHA benchmark, containing 50 episodes and 25,000 frames with dual-arm state and action vectors at 50 Hz.
50Reported episodes
50 HzReported frequency
8m 20sReported duration
LeRobot v2Dataset format
KNOWN CONFIGURATION
Robot, observations and reuse
Robot or capture systemALOHA dual-arm (simulation)
Reported observationsTop RGB, 14D state, Bimanual actions
ActionsRobot actions are included
EnvironmentSimulation · bimanual
Licence shown on source cardMIT
TRUST & LIMITATIONS
What EmbodiedData has—and has not—checked
Original Hugging Face source linkChecked
Summary against publisher cardReviewed
File integrity and missing valuesNot independently tested
Privacy, safety and legal complianceNot independently tested
Always read the original dataset card, licence, limitations and file documentation before training or redistribution.
SOURCE & CITATION
Use the publisher’s citation
EmbodiedData does not invent a citation or dataset version. Open the original card and copy the citation requested by the publisher.
Open original dataset card ↗