Provides image understanding capabilities by analyzing single or multiple images (local files, URLs, or base64) via MiniMax VL API, complementing text-only models.
