This is your work, valued

Autodistill

Expert
@autodistill

Use big, slow foundation vision models to train smaller, faster models.