Skip to main content

SE3 Labs · Developer resources

Your robot policy.
On real hardware.

Connect your model to SE3 stations, run evaluations, and inspect the results. Start with a guided tutorial or look up a Python module in the package reference.

How evaluation works

  1. Connect your policyRun your model beside the SDK.
  2. Evaluate on a stationExchange observations and joint actions.
  3. Review the outcomeInspect episode results and operator scores.
Setup and reference

Documentation

Install the SDK, then browse se3labs.sdk clients and commands and se3labs.interface policy contracts and data types.

Read the documentation Learn by doing

Tutorials

Evaluate your policy through the Python API, XPolicyLab, or an agent using ACT Harness. Try a complete example that moves the robot arms left and right.

Start a tutorial