mulligan2026apache-2.0
real-marker-d2-r02-eval-b02
Evaluation dataset of 100 episodes from the Mulligan robotics benchmark, containing real robot marker-based manipulation task recordings with multi-camera observations collected at 15 FPS.
Downloads44
Episodes100
Why This Matters for Physical AI
This evaluation dataset enables benchmarking of robot manipulation policies on real hardware, supporting policy development and comparative evaluation through the Mulligan framework.
Technical Profile
- Modalities
- rgb
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 100
- Data Format
- parquet
- Annotation Types
- language_instructions
- License
- apache-2.0
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack