Generate depth maps from images and videos
Detect objects in images or videos
Generate chat responses using Llama-2 13B model