A monocular depth estimation model using a Vision Transformer (ViT-L/14) backbone from the original Depth Anything release.