HN 提问：大型语言模型的下一步是什么？
1 分•作者: apatheticonion•2 天前
我一直在将 DeepSeek v4 flash 用于引导式代理工作流（IDE 内、提示/审查差异），并且在这些工作流中没有发现使用 Haiku、Opus、Sonnet 有任何明显的好处。
每天花费大约一美元，我的生产力就能提高一倍多。那么，前沿模型是什么？它们在解决什么问题？
它们是为一次性应用设计的吗？我们是否期望能够“ YOLO 式”地提示 LLM，让它们在无人监督的情况下完成工作？
查看原文
I've been using DeepSeek v4 flash for guided agent workflows (in-IDE, prompt/review diffs) and have found no noticeable benefit to using Haiku, Opus, Sonnet in this workflow.<p>For around a dollar a day, I am able to more than double my own productivity. What are the frontier models for/what are they trying to solve?<p>Are they designed to one shot applications? Is the expectation that we want to be able to "yolo" prompt LLMs and have them complete work unsupervised?