今天 Claude 号被封了,把 claude Code 模型换成了 deepseek,又写代码又调研,感觉还真不错。推理质量不像 claude opus 那么机灵,但还是靠谱的。而且太省钱了,充了 10 块钱用 thinking 模型能玩好几天。之前试过 GLM 5.1 太慢了,minimaxi 2.7 推理质量太差了
I put a lot of heart into my technical writing, I hope it's useful to you all.
📌 Here's a pinned thread of everything I've written.
(much of this will be posted on the Claude blog soon as well)