[Domestic Large Models Dominate Global Rankings for 5 Consecutive Weeks] The domestic large model MiniMax M2.5 has topped the global large model usage rankings for five consecutive weeks. In the company headquarters' office area, product development personnel demonstrated the stark price difference to reporters. "For overseas models with the same capability level, the price could be ten times higher compared to ours." Cost-effectiveness is becoming the core competitive advantage attracting global users to domestic large models. Where does this cost-effectiveness come from? The first key factor is technological innovation. By innovating at the foundational architecture level, inference costs have been fundamentally reduced. Simply put, the same tasks can be completed with fewer tokens. Another critical factor is energy advantages. Industry insiders told reporters that electricity costs currently account for as much as 70% to 80% of computing power costs. For AI clusters requiring millions of GPUs to operate in parallel, even a slight difference in electricity prices can translate into astronomical figures in annual operating costs. Stable energy supply and relatively low electricity prices have built a more competitive cost barrier for China's AI industry. (CCTV Finance)