SmolVLA
A compact, efficient VLA trained on community LeRobot datasets.
SmolVLA is a 450M-parameter vision-language-action model that runs on consumer GPUs and CPUs, released as part of LeRobot with asynchronous inference.
Description
SmolVLA is a 450M-parameter vision-language-action model that runs on consumer GPUs and CPUs, released as part of LeRobot with asynchronous inference.