MiniMax M3.1 Flash

A fast language model optimized for autonomous agents and coding tasks. It reaches generation speeds exceeding two hundred tokens per second while keeping costs low.