MiniMax Launches M3.1 Flash: High-Speed Agentic Model Slashing Costs by 70%
ai••2 min•minimax.io
Chinese AI lab MiniMax has debuted M3.1 Flash, an ultra-fast foundation model engineered for agentic loops, coding workflows, and computer use, delivering 3x faster inference throughput, 70% lower operational costs, and an upcoming open-weights release.
Read full article→#MiniMax