Skip to content

Research ยท October 2026 edition

Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License

34 open datasets of robot, first-person video, driving and drone training data were released from 2023 on; each is listed with who released it, its stated size, how its data was collected and its licence, linked to its source.

Read the month's story: Robot Training Jobs in October 2026: The Month in Data

Not final yet. The month has ended, yet stragglers in the data may still land up to 3 Nov 2026; after that the edition is locked permanently.

Covers
2 Jan 2023 to 28 Oct 2026
Dated by when each event occurred
Calculated
1 Nov 2026, 06:05 UTC
Settles when locked
Where it comes from
Each dataset's paper, project page or dataset card, and reporting where the release gives no date or worker count, each linked with a verbatim quote kept in the dataset.
Method
Revision 1. Terms defined
CC BY 4.0, citing this report

Each figure here comes as CSV and JSON files, emailed to your work address.

Need robot training data collected? We run the project: teleoperation sessions, first-person video, filmed household tasks or labeled robot video, staffed from the people who apply for this work through us, checked against your standard, starting with a paid pilot.

How we run projects

What the data shows

  1. 1

    34 open datasets since 2023: 16 robot manipulation, 12 first-person video, 4 driving and 2 drones.

    Figure 2
  2. 2

    The largest by stated hours is Egocentric-100K (Build AI, 19 Dec 2025): 100,405 hours.

    Figure 3
  3. 3

    12 of the 34 were collected by people teleoperating robots, the work this site's teleoperation listings pay for.

    Figure 1
  4. 4

    The newest release on record: EMG-1 (20 Oct 2026). Encord released EMG-1, which it calls the largest open-source dataset pairing muscle signals to first-person video, so that robot models can learn what hands do when a camera cannot see them; it says the dataset is over 30 times larger than the next-largest public dataset of its kind.

    Figure 1

Figure 1

Open training datasets released since 2023

Which public datasets teach robots, embodied AI, self-driving systems and drones, who released them, how big are they and how was the data collected?

DatasetReleased byKindCollected by
EMG-1PR Newswire (Encord)Encord, with Wearable DevicesFirst-person videoWearable camera300n/a20 Oct 2026
Open-AoEHugging Face (Open-AoE dataset card)Ant GroupFirst-person videoWearable camera2,000n/a17 Oct 2026
EgoSuite-Open100KLightwheel (Hugging Face blog)Lightwheel with Hugging FaceFirst-person videoWearable camera10,000n/a26 Aug 2026
BEHAVIOR Challenge 2026 demonstrationsStanford Vision and Learning LabStanford Vision and Learning Lab (BEHAVIOR)Robot manipulationTeleoperation1,95020,0002 Jul 2026
HIW-500 (Humanoids-in-the-Wild 500)BitRobot (Hugging Face)BitRobot Network with Hugging Face and Unitree RoboticsRobot manipulationTeleoperation50023,00025 Jun 2026
ABC-130KarXiv (UC Berkeley, MIT, Amazon FAR, XDOF, CMU)XDOF with UC Berkeley, MIT, Amazon FAR and Carnegie Mellon UniversityRobot manipulationTeleoperation3,500130,00025 Jun 2026
MolmoAct 2-Bimanual YAMAi2Ai2 (Allen Institute for AI)Robot manipulationNot stated or other72032,2465 May 2026
Egocentric-100KBuild AI (Hugging Face)Build AIFirst-person videoWearable camera100,405n/a19 Dec 2025
RoboCOINarXiv (Beijing Academy of Artificial Intelligence)Beijing Academy of Artificial Intelligence (BAAI) and partner institutionsRobot manipulationTeleoperationn/a180,00021 Nov 2025
Egocentric-10KBuild AI (Hugging Face)Build AIFirst-person videoWearable camera10,000n/a11 Nov 2025
Waymo Open Dataset for End-to-End Driving (WOD-E2E)arXiv (Waymo)WaymoDrivingNot stated or other12n/a30 Oct 2025
Humanoid EverydayarXiv (USC, Toyota Research Institute)University of Southern California and Toyota Research InstituteRobot manipulationTeleoperationn/a10,3009 Oct 2025
Galaxea Open-World DatasetarXiv (Galaxea AI)Galaxea AIRobot manipulationTeleoperation500100,00030 Aug 2025
MolmoAct DatasetAi2Ai2 (Allen Institute for AI)Robot manipulationNot stated or other2210,00012 Aug 2025
LeRobot community datasets (SmolVLA pretraining set)Hugging FaceHugging Face (LeRobot) and community contributorsRobot manipulationMixedn/an/a3 Jun 2025
UAV-FlowarXiv (Beihang University, NUS)Beihang University and National University of SingaporeDronesTeleoperationn/a30,00021 May 2025
EgoDexarXiv (Apple)AppleFirst-person videoWearable camera829338,00016 May 2025
Fourier ActionNetFourier (ActionNet project page)FourierRobot manipulationTeleoperation14030,00018 Mar 2025
HD-EPICarXiv (University of Bristol and partners)University of Bristol, Leiden University, Singapore Management University and University of Bath (EPIC-KITCHENS team)First-person videoWearable camera41n/a6 Feb 2025
AgiBot WorldarXiv (AgiBot World Colosseo team)AgiBotRobot manipulationTeleoperation2,9761,001,55230 Dec 2024
RoboMINDarXiv (Beijing Innovation Center of Humanoid Robotics, Peking University)Beijing Innovation Center of Humanoid Robotics (X-Humanoid), Peking University and Beijing Academy of Artificial IntelligenceRobot manipulationTeleoperation306107,00018 Dec 2024
MAN TruckScenesarXiv (MAN Truck & Bus, TUM)MAN Truck & Bus with the Technical University of MunichDrivingNot stated or othern/an/a10 Jul 2024
CityNavarXivTokyo Institute of Technology, The University of Tokyo, Kyoto University and partnersDronesNot stated or othern/a32,63720 Jun 2024
NymeriaarXiv (Meta Reality Labs Research)Meta (Reality Labs Research, Project Aria)First-person videoWearable camera300n/a14 Jun 2024
HOT3DarXiv (Meta)Meta (Reality Labs)First-person videoWearable cameran/an/a13 Jun 2024
DROID (Distributed Robot Interaction Dataset)arXiv (DROID collaboration)DROID collaboration (13 institutions)Robot manipulationTeleoperation35076,00019 Mar 2024
Aria Everyday Activities (AEA)arXiv (Meta)Meta (Project Aria)First-person videoWearable camera8n/a20 Feb 2024
Ego-Exo4DarXiv (Meta FAIR and university partners)Meta (FAIR and Project Aria) with 14 university partnersFirst-person videoMixed1,286n/a30 Nov 2023
Open X-EmbodimentarXiv (Open X-Embodiment Collaboration)Google DeepMind and the Open X-Embodiment Collaboration (21 institutions)Robot manipulationMixedn/a1,000,0003 Oct 2023
BridgeData V2arXiv (UC Berkeley, Stanford)UC Berkeley (RAIL) with Stanford and collaboratorsRobot manipulationMixedn/a60,09624 Aug 2023
RH20TarXiv (Shanghai Jiao Tong University)Shanghai Jiao Tong UniversityRobot manipulationTeleoperationn/a110,0002 Jul 2023
Zenseact Open Dataset (ZOD)arXiv (Zenseact)ZenseactDrivingNot stated or othern/an/a3 May 2023
Ego4D (version 2)Ego4DMeta (FAIR) and the Ego4D university consortiumFirst-person videoWearable camera3,670n/a1 Feb 2023
Argoverse 2arXiv (Argo AI)Argo AIDrivingNot stated or othern/an/a2 Jan 2023

Base: 34 datasets released from 2023 to 28 Oct 2026, each linked to its paper or release page (reviewed 28 Oct 2026).

Hours and episodes are as each release states them; a blank means the release gives no figure. Episodes, trajectories and demonstrations are counted as one.

Credit: compiled by RobotTraining.jobs Research, Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition.

Figure 2

Datasets by what they teach

How many open datasets are there for each kind of robot learning, and how was their data gathered?

KindDatasetsTeleoperatedWearable camera
Robot manipulation
16
110ABC-130K (3,500 h)
First-person video
12
011Egocentric-100K (100,405 h)
Driving
4
00Waymo Open Dataset for End-to-End Driving (WOD-E2E) (12 h)
Drones
2
10n/a

Base: 34 datasets.

Collected by: how the release says the data was gathered. Teleoperation means people steering robots remotely or in VR; wearable camera means people filming their own hands and tasks.

Credit: compiled by RobotTraining.jobs Research, Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition.

Figure 3

The largest datasets by hours of recorded data

Which open datasets hold the most hours of human-made robot training data?

DatasetHours
Egocentric-100KBuild AI (Hugging Face)Wearable camera14,228 factory workers (as reported from the dataset's technical details)
EgoSuite-Open100KLightwheel (Hugging Face blog)Wearable cameraA globally distributed workforce of tens of thousands of collectors
Egocentric-10KBuild AI (Hugging Face)Wearable camera2,153 factory workers (as reported), in factories numbered 001 to 085 on the dataset card
Ego4D (version 2)Ego4DWearable camera923 unique participants in 74 locations in 9 countries; 88 researchers in an international consortium
ABC-130KarXiv (UC Berkeley, MIT, Amazon FAR, XDOF, CMU)Teleoperationn/a
AgiBot WorldarXiv (AgiBot World Colosseo team)TeleoperationSkilled teleoperators working more than 100 homogeneous robots in a data collection facility of over 4,000 square meters
Open-AoEHugging Face (Open-AoE dataset card)Wearable camera500+ contributors using 400+ smartphones in natural environments
BEHAVIOR Challenge 2026 demonstrationsStanford Vision and Learning LabTeleoperationTeleoperation data in simulation provided by Simovation
Ego-Exo4DarXiv (Meta FAIR and university partners)Mixed740 participants (camera wearers) in 13 cities; Meta's FAIR, Project Aria and 14 university partners
EgoDexarXiv (Apple)Wearable cameran/a
MolmoAct 2-Bimanual YAMAi2Not stated or otherCurated with support from Cortex AI
HIW-500 (Humanoids-in-the-Wild 500)BitRobot (Hugging Face)TeleoperationHuman teleoperators in 12 homes across Southeast Asia
Galaxea Open-World DatasetarXiv (Galaxea AI)Teleoperationn/a
DROID (Distributed Robot Interaction Dataset)arXiv (DROID collaboration)Teleoperation50 data collectors in North America, Asia and Europe at 13 institutions, over 12 months
RoboMINDarXiv (Beijing Innovation Center of Humanoid Robotics, Peking University)Teleoperationn/a
EMG-1PR Newswire (Encord)Wearable cameraPeople wearing Wearable Devices wristbands that read forearm muscle signals while completing real-world tasks on first-person video
NymeriaarXiv (Meta Reality Labs Research)Wearable camera264 participants across 50 locations
Fourier ActionNetFourier (ActionNet project page)Teleoperationn/a
HD-EPICarXiv (University of Bristol and partners)Wearable camera9 participants recording in their own kitchens for at least 3 consecutive days
MolmoAct DatasetAi2Not stated or otherCollected in-house by Ai2 over about 3 months
Waymo Open Dataset for End-to-End Driving (WOD-E2E)arXiv (Waymo)Not stated or othern/a
Aria Everyday Activities (AEA)arXiv (Meta)Wearable cameraMultiple wearers (1 to 2 per sequence) in five geographically diverse indoor locations

Base: 22 of 34 datasets state their hours.

Hours are not totalled across datasets: releases count them differently (raw video, cleaned episodes, approved data).

Credit: compiled by RobotTraining.jobs Research, Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition.

Figure 4

Size, contributors and licence of each dataset

What does each dataset contain, who recorded it and under what licence can it be used?

DatasetSize as stated
EMG-1PR Newswire (Encord)more than 300 hours of synchronized muscle-signal (EMG) and first-person video recordings during real-world tasks, alongside optical motion captureNot stated on the release page
Open-AoEHugging Face (Open-AoE dataset card)approximately 2,000 hours of manipulation video with MANO hand poses, camera trajectories and temporally localized atomic-action annotationsOpen-AoE Dataset License (research and commercial use with attribution; MANO-derived files also under the MANO licence)
EgoSuite-Open100KLightwheel (Hugging Face blog)100,000 hours planned; first 10,000 hours available; 15,000+ tasks and 15,000+ real-world collection scenescommercial-training-no-resale-v1.0 (academic research and commercial training)
BEHAVIOR Challenge 2026 demonstrationsStanford Vision and Learning Lab20,000 human teleoperation demos, 1,950 hours in total (in simulation)MIT
HIW-500 (Humanoids-in-the-Wild 500)BitRobot (Hugging Face)500+ hours, 23K+ episodes and 10+ TB of raw data from 12 real homesCC BY 4.0
ABC-130KarXiv (UC Berkeley, MIT, Amazon FAR, XDOF, CMU)3,500 hours over 130K episodes across 195 tasks (project page: 134,806 episodes, 3,553 hours), plus 400 hours of sim-teleop dataApache 2.0
MolmoAct 2-Bimanual YAMAi2over 720 hours of training demonstrations (the blog also calls it a 700-hour collection); 32,246 episodes on the Hugging Face cardApache 2.0
Egocentric-100KBuild AI (Hugging Face)100,405 hours, 10.8 billion frames, 2,010,759 clips, 24.79 TBApache 2.0
RoboCOINarXiv (Beijing Academy of Artificial Intelligence)over 180,000 demonstrations from 15 robotic platforms, 421 bimanual tasks, 432 objects, 16 environmentsNot stated on the release page
Egocentric-10KBuild AI (Hugging Face)10,000 hours, 1.08 billion frames, 192,900 clips, 16.4 TBApache 2.0
Waymo Open Dataset for End-to-End Driving (WOD-E2E)arXiv (Waymo)4,021 driving segments (approximately 12 hours) of 20 seconds each, curated for long-tail scenariosWaymo Dataset License Agreement for Non-Commercial Use
Humanoid EverydayarXiv (USC, Toyota Research Institute)10.3k trajectories and over 3 million frames across 260 tasks in 7 categoriesNot stated on the release page
Galaxea Open-World DatasetarXiv (Galaxea AI)500 hours, 100K demonstration trajectories, more than 150 tasks across 50 real-world scenesCC BY-NC-SA 4.0
MolmoAct DatasetAi2~10,000 distinct robot episodes (the MolmoAct 2 announcement describes 22 hours, about 10.6K trajectories)CC BY 4.0
LeRobot community datasets (SmolVLA pretraining set)Hugging Face487 curated community datasets focused on the SO100 robot arm, standardized at 30 FPS, around 10 million frames (the SmolVLA paper counts 481 datasets)Varies by dataset (compatibly licensed, open-source community datasets)
UAV-FlowarXiv (Beihang University, NUS)real-world dataset with 30k flight trajectories, plus a simulation datasetNot stated on the release page
EgoDexarXiv (Apple)829 hours of 1080p, 30 Hz egocentric video with 338,000 episodes across 194 tasks (90 million frames)CC BY-NC-ND
Fourier ActionNetFourier (ActionNet project page)over 30K teleoperated trajectories or around 140 hours of interaction dataCC BY-NC-SA 4.0 (dataset); Apache 2.0 (training and evaluation tools)
HD-EPICarXiv (University of Bristol and partners)41 hours of video in 9 kitchens, 69 recipes, 59K fine-grained actions, 51K audio events, 20K object movementsNot stated on the release page
AgiBot WorldarXiv (AgiBot World Colosseo team)1,001,552 trajectories totalling 2,976.4 hours, 217 tasks, 87 skills and 106 scenes (latest version per the paper); the GitHub Beta release lists 1,003,672 trajectories (~43.8 TB)CC BY-NC-SA 4.0
RoboMINDarXiv (Beijing Innovation Center of Humanoid Robotics, Peking University)107k demonstration trajectories (305.5 hours) across 479 tasks and 96 object classes on 4 robot embodiments, plus 5k real-world failure demonstrationsApache 2.0 (Hugging Face listing)
MAN TruckScenesarXiv (MAN Truck & Bus, TUM)more than 740 scenes of 20 s each, with 4 cameras, 6 lidar and 6 radar sensorsCC BY-NC-SA 4.0
CityNavarXiv32,637 human demonstration trajectories paired with language descriptions, over 4.65 km2 of Cambridge and BirminghamNot stated on the release page
NymeriaarXiv (Meta Reality Labs Research)300 hours of daily activities from 264 participants across 50 locations, over 399 km travelledNymeria dataset license (Meta's own terms, in the repository)
HOT3DarXiv (Meta)over 833 minutes (more than 3.7M images) of multi-view image streams, 19 subjects, 33 rigid objectsHOT3D license agreement (Meta)
DROID (Distributed Robot Interaction Dataset)arXiv (DROID collaboration)76k demonstration trajectories or 350 hours of interaction data, across 564 scenes and 84 tasks (the project page says 86 tasks)CC BY 4.0
Aria Everyday Activities (AEA)arXiv (Meta)143 daily activity sequences, over 7.5 accumulated hours, recorded in 5 locationsNot stated on the release page
Ego-Exo4DarXiv (Meta FAIR and university partners)1,286 hours of video combined (1,286.3 ego and exo hours) from 740 participants in 13 cities and 123 scene contextsEgo-Exo4D license agreement (must be signed and approved before download)
Open X-EmbodimentarXiv (Open X-Embodiment Collaboration)1M+ real robot trajectories spanning 22 robot embodiments, pooled from 60 existing robot datasets; 527 skills (160,266 tasks)CC BY 4.0 for data and materials, Apache 2.0 for software
BridgeData V2arXiv (UC Berkeley, Stanford)60,096 trajectories across 24 environments on a WidowX 250 arm: 50,365 teleoperated demonstrations and 9,731 rollouts from a scripted pick-and-place policyCC BY 4.0
RH20TarXiv (Shanghai Jiao Tong University)over 110,000 contact-rich robot manipulation sequences, each paired with a human demonstration video; tasks drawn from RLBench (48), MetaWorld (29) and 70 self-proposed tasksMixed: RH20T-C subset CC BY-SA 4.0, RH20T-NC subset CC BY-NC 4.0
Zenseact Open Dataset (ZOD)arXiv (Zenseact)100k curated camera frames, 1,473 sequences of 20 seconds and 29 drives of a few minutes, collected over two years in 14 European countriesCC BY-SA 4.0
Ego4D (version 2)Ego4D3,670 hours of daily-life video from 923 participants in 74 locations across 9 countriesEgo4D license agreement (must be accepted and approved before download)
Argoverse 2arXiv (Argo AI)Sensor Dataset: 1,000 annotated sequences; Lidar Dataset: 20,000 sequences; Motion Forecasting Dataset: 250,000 scenarios; six US citiesCC BY-NC-SA 4.0

Base: All 34 datasets.

A licence is summarised as the release page gives it; read the licence itself before using a dataset.

Credit: compiled by RobotTraining.jobs Research, Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition.

Robotics labs ยท Fleet operators ยท Data vendors ยท Research teams

Buy the data behind this report

This report sums up records we gather ourselves, day after day:

  • 1,432 robot training roles followed, history included

    Every role on 100 employer job boards, read each day since 12 Sep 2026: its pay, its place, and the first and last day it was up.

  • 536 apply clicks over 28 days

    Which robot training roles people really apply to, by kind of work, pay and place. The clicks happen here, so no one else can count them.

  • 25 sourced industry datasets

    Funding, revenue, deals, fleets, deployments and prices across the robot training industry, each record linked to the publication it comes from.

On offer

  • The raw records

    Every robot training role we track: title, employer, place, kind of work, pay as stated, and the days it was first and last open.

  • A regular feed

    New and closed roles every week or month, as file drops or through the data endpoints.

  • A custom cut

    Any figure here by country, city, kind of work, vehicle, employer or date, or your own hiring benchmarked against the market.

Need the people behind this data for your own robots? (you contract them yourself) or (we deliver the robot data).

Investing in robotics, AV or robot-data companies? The Robot hiring radar is these records role by role, every day, as files.

Robot hiring brief: free, once a month

The robot-training job market in figures, on the first weekday of each month: open roles, what opened and closed, who added and who cut, and posted pay by kind of work. To your work email.

How it was measured

Each dataset's paper, project page or dataset card, and reporting where the release gives no date or worker count, each linked with a verbatim quote kept in the dataset. Each row carries a link to its publication and gives the number exactly as published, with no estimates or conversions. A note under every figure says what it covers.

Episode
One recorded attempt at a task (also called a trajectory or demonstration).
Egocentric
First-person video filmed from a camera the person wears, usually on the head.

Limits of this data

  • Built only from public sources. Anything never announced or reported is absent, and private-company figures are whatever the company or the press said, unaudited.
  • Releases count hours and episodes differently, so sizes are compared only as each release states them.
  • Datasets that are mostly synthetic, or released before 2023, are left out.

Collection and classification of listings are explained on the methodology page. Freshness, completeness and consistency are tested daily in the data quality report.

Monthly editions

The live page shows the dataset as of its latest review, and new events join it once they are made public. Every finished month also gets its own edition, locked two days after month end, so a number quoted from it stays fixed.

  • Live pageLatest review
  • October 2026 edition (you are here)Will lock when the next daily refresh runs

Quoting this report

You may quote, chart or republish these figures under CC BY 4.0, provided you link back to this page. This edition is locked: its numbers will not change.

Citation

RobotTraining.jobs (2026). Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition. https://robottraining.jobs/research/open-robot-training-datasets/2026-10

BibTeX

@misc{robottrainingjobs_open_robot_training_datasets_2026_10,
  author = {{RobotTraining.jobs}},
  title = {Open-Source Robotics Datasets: Robot Learning and Training Datasets by Size, Method and License, October 2026 edition},
  year = {2026},
  howpublished = {\url{https://robottraining.jobs/research/open-robot-training-datasets/2026-10}},
}
Files: every figure in one CSV (a row per value) or as JSON, emailed to your work address.Other open data: datasets and API

Keep reading

Browse all 91 reports