📄 中文摘要
MFS Corp 在构建自主 AI 驱动的组织时，面临着早期基础设施选择的挑战，尤其是在本地 LLM 和云 API 之间的选择。经过经济分析，云端解决方案的成本高达每月 600-800 美元，而采用本地 LLM 的混合方案则可以在日常任务中进行本地推理，仅在战略决策时使用云服务，从而显著降低成本。最终，MFS Corp 选择了结合本地和云端的混合模式，以实现更高的经济效益和灵活性。
为何我们选择本地 LLM 而非仅云端解决方案（以及何时打破这一规则）
📄 English Summary
Why We Chose Local LLMs Over Cloud-Only (and When We Break That Rule)
Building MFS Corp as an autonomous AI-driven organization involved making challenging infrastructure decisions, particularly between local LLMs and cloud APIs. Economic analysis revealed that a cloud-only scenario could cost between $600 and $800 monthly. In contrast, a hybrid approach utilizing local LLMs for routine tasks while reserving cloud services for strategic decisions significantly reduced costs. Ultimately, MFS Corp opted for a combination of local and cloud solutions to achieve greater economic efficiency and flexibility.