minimax-m3.1POST
开发者:Kimi K3、GLM 5.3、DeepSeek V4.1、M3.1 已可互换日常使用
开发者称这几个模型在日常任务上已达前沿水平,Claude/GPT 额度用尽时直接换 M3.1 等继续干活,配合自建记忆工具。
Fabio Makita
@AkitaOnRails4 天前收录原推发布于 2026.10.24
Fabio Makita @AkitaOnRails原推发布于 2026.10.24
Actually Kimi K3, GLM 5.3, DeepSeek v4.1, even MiniMax M3.1 are already "frontier" level for most day to day tasks (maybe not for nice advanced research and such, but I don't need those). You can use them interchangeably. Every time claude or gpt cut me off on damn quotas, I switch to any of the aforementioned models and keep going normally. (helped by my http://aimemory.sh)
这条是引用着另一条写的:x.com/theahmadosman/status/2113743781132669067
x.com/akitaonrails/status/2113805047611961713采集于 2026.10.24 12:10