Capture • Dataset • Evaluation

EgoVerse Human data from around the world, built for robot learning

EgoVerse is an ecosystem for curating, accessing, and learning from human data for robot learning. It hosts a "living" dataset, continuously expanded by the consortium, and driven by a research community advancing rigorous studies of human-to-robot transfer across tasks and embodiments.

Dataset browser → Install + Code The Consortium

Read the EgoVerse Paper →

Replace the hero video with a “coalition” montage that shows cross-lab and industry capture setups, plus robot rollouts.

Coalition montage Cross-lab capture Robot evaluation

Real-World Evaluations

By Partner Institutes

Georgia Tech (RL2) Task: object-in-container • ID/OOD
Stanford University (REAL Lab) Task: cup-on-saucer • bimanual precision
UC San Diego (Wang Lab) Task: bag-grocery • long-horizon

If you want 6–9 tiles, duplicate the blocks. Keep a fixed aspect ratio and consistent titles.

Global Data Collection

Data from around the world, continuously growing.

Dataset Snapshot

1,362
Hours of human demos
~80k
Episodes
1,965
Tasks
240
Scenes
2,087
Unique demonstrators

Stats from the current release.