smolvla_base is an open-source vision-language-action model developed as part of the LeRobot project. It processes visual inputs and language instructions to output actions suitable for robotic control. The model is designed for imitation learning tasks and can be used with the LeRobot library for training and inference on real or simulated robots.
In the Other AI space, Smolvla Base takes a focused approach. It focuses on training compact models that can understand visual scenes and output robot actions. It is built as an open-source project for robotics researchers and developers. The project is open source (Apache-2.0). It ships for the web and API.
LeRobot builds and maintains Smolvla Base, and it first shipped in 2024. The project is developed in the open on GitHub with 26.3k stars and 248 commits in the last 90 days. Among its 3 catalogued features are vision-Language-Action, Imitation Learning, and Robot Control.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do