Powerful multimodal AI model from MiniMax supporting text, image, audio, and video understanding. Competitive with leading Western models for diverse AI applications.