构建自己的评估与运行环境：为什么企业应该掌握自己的智能栈
Own your intelligence stack, folks! I will never get tired of saying this. Your business's very exi...
Rauch 和 Elvis 建议企业别全押闭源 API，先从自建 evals 和 harness 入手，逐步掌握数据、运行时和智能体，保留换模型的选择权。
Guillermo Rauch 与 Elvis Saravia 在 X 上讨论企业自建智能栈的必要性，观点是依赖外部 API 的企业在智能时代面临生存风险。建议从评估（evals）入手，先建自己的 harness，再逐步向栈的下层推进。文中承认开源模型目前运行和微调成本仍然很高，大规模本地部署尚不可行。自建 harness 的价值在于保留可选性，并帮助企业测量下一阶段该在哪一层投入。
Own your intelligence stack, folks! I will never get tired of saying this. Your business's very exi...
Own your intelligence stack, folks! I will never get tired of saying this. Your business's very existence is at play here. But you don't have to do this all at once if you can't afford it. I mean, open models are still very expensive to run and tune. If you want to start somewhere, I suggest the harness. Build your evals and quickly move down the stack. The great part about building your own harness is that it gives you optionality, unlocks capabilities, and helps you measure where else in the intelligence stack you need to invest next. So while open-source was an alternative in the old era, today it is how you survive and thrive in the age of intelligence. Guillermo Rauch @rauchg Own your intelligence. Anyone that opposes this has something to sell you (perhaps your own intelligence back to you.) Local models need to continue proliferating. The “absolute unit” massive models may be out of reach for now locally, so your best option is to, indeed, keep your options open. Own your harness, data, connections, runtime: your agents. Embrace open source. 🔗 View Quoted Tweet 💬 67 🔄 7 ❤️ 118 👀 11136 📊 63 ⚡
