CALVIN
Canonical Summary
Freiburg long-horizon benchmark: 4 indoor environments, about 24 hours of play data, and 20K language directives.
Lead Organization
University of Freiburg
Key Technical Metric
4 envs; 20K directives
Maturity / Status
Public benchmark
Governance / License
Open-source simulated benchmark
Technical Specifications
Environments 4 indoor manipulation environments
Play data About 24 hours of teleoperated unstructured play
Language 20K language directives
Control 7-DOF continuous control
Zero-shot split Train on three environments, test on an unseen scene
Verification: verified · Retrieved: 2026-10-19 ·
Primary Source Reference