Analyze an image URL for faces, scene details, and embeddings
Generate text embeddings for your input