act_policy_test4 is an open-source imitation-learning policy model designed for robotics applications, focusing on predicting short action chunks from teleoperated data. It can be integrated with LeRobot and other frameworks for training, evaluation, and deployment in robotics workflows. Ideal for robotics researchers and developers seeking ready-to-use policy models.
In the Other AI space, Act Policy Test4 takes a focused approach. It enables robotics developers to implement and evaluate imitation-learning policies for action chunking without building models from scratch. It is built as an open-source project for robotics researchers and developers. Act Policy Test4 is open source under the Apache-2.0 license. The product ships for the web, API, and the command line.
Hilman09 builds and maintains Act Policy Test4, and the product first shipped in 2024. Development happens publicly on GitHub with 25.8k stars and 156 commits in the last 90 days. PulseGate's similarity index finds few close equivalents — Act Policy Test4 occupies a relatively distinct niche. Key capabilities include imitation learning, action chunking, and robotics policy.
Latest indexed changes and source events
Hilman09/act_policy_test4 discovered by the PulseGate indexer
