SmolVLA
Hugging Face / LeRobot
A smaller VLA reference with a documented fine-tuning workflow.
What you can explore
Multiple camera views, robot state and language conditioning. Continuous action chunks, learned with flow matching.
Before you use it
The documentation illustrates SO100/SO101 arm tasks. Those examples do not establish independent finger control.
Original contributors
Hugging Face / LeRobot
License / access: Base model card lists Apache-2.0. Check each dataset and dependency independently.
Keep exploring
LeRobot
Tools for recording datasets, training policies and exploring robot-learning workflows.
ACT / Action Chunking with Transformers
An imitation-learning implementation that predicts chunks of robot actions from demonstrations.
Editorial reference · checked 2026-09-25