Benchmark RadarAXIS
AXIS is a community-driven data engine and benchmark for robot manipulation, providing browser-based teleoperation for data collection, automated task generation and validation, and a dataset of 207 tasks and 50K+ trajectories, with a systematic held-out protocol for policy evaluation.
- Released
- 2026-07-23
- Readiness
- Inspectable
- Primary field
- Robotics & Autonomous Systems
Why it matters
Scaling robot learning requires diverse data and standardized evaluation; AXIS offers a growable pipeline and unified evaluation suite, enabling comparison of VLA policies and studying scaling behavior.
Motivation
Learning effective robot manipulation policies requires diverse, high-quality demonstrations, yet existing data pipelines are often difficult to scale because they rely on specialized hardware, centralized operators, or fixed task suites.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.