Datasetroboticsv1.0

Open X-Embodiment

by Google DeepMind / Consortium · free · Last verified 2026-03-17

Open X-Embodiment (OXE) is a massive robotics dataset combining over 1 million demonstration episodes from 22 distinct robot embodiments. It covers 527 skills and is designed to train generalist robot policies that can transfer skills across diverse hardware, serving as a key resource for vision-language-action models.

https://robotics-transformer-x.github.io ↗

B—Above Average

Adoption: B+Quality: AFreshness: ACitations: AEngagement: F

Specifications

License: Apache 2.0
Pricing: free
Capabilities: Large-scale imitation learning, Cross-embodiment policy training, Vision-Language-Action (VLA) model development, Zero-shot skill transfer to new robots, Benchmarking for generalist robot policies, Training foundation models for robotics, Data aggregation from heterogeneous sources, Real-world robotic manipulation learning
Integrations
Use Cases: [object Object], [object Object], [object Object], [object Object]
API Available: No
Tags: robotics, manipulation, multi-robot, imitation-learning, embodied-ai, dataset, robot-learning, vla-model, transfer-learning, foundation-model
Added: 2026-03-17
Completeness: 0.9%

Index Score

66.1

Adoption

Quality

Freshness

Citations

Engagement

Need this tool deployed for your team?

Get a Custom Setup

Explore the full AI ecosystem on Agents as a Service