A simulated Franka arm on 40 tasks in four suites: spatial layout, objects, goals and long multi-step tasks. Success rate in %, fine-tuned on each suite.
Best score in each column highlighted. Select a column to sort.
Reported, not re-run. Numbers come from papers and model docs. Setups differ in cameras, training data and number of trials.
Simulation is a starting point. High LIBERO scores are common now; real robots and new scenes are harder. Test on your task in the playground.
{{bAbout}}
{{tableNote}}
University labs run pairs of policies on the same task on DROID setups (Franka arm, wrist and side cameras) without knowing which is which, then pick the better run. In the paper's pool of seven DROID policies, π0-FAST-DROID was the strongest.
Read the paper ↗Every routed run on a pilot robot is logged and scored. Once pilots are live, results from real factory tasks will appear here next to the lab benchmarks.
Reported, not re-run. Numbers come from papers and model docs. Setups differ in cameras, training data and number of trials.
Simulation is a starting point. High LIBERO scores are common now; real robots and new scenes are harder. Test on your task in the playground.