网友质疑 DeepSeek V4.1-1017 提升来自 OpenAI 蒸馏的说法
一条关于 DeepSeek V4.1-1017 提升来源的争论帖,作者拿 Luna 没跑赢 V4.1 来反驳蒸馏说,观点挺有意思。
一条 X 帖子讨论 DeepSeek V4.1-1017 相对 V4.1 的性能提升来源。发帖人反驳了提升主要靠蒸馏 OpenAI 模型的说法,理由是 OpenAI 自己的小模型 Luna 并未超过 V4.1。他还指出 Luna 速度更慢、规模可能更小,质疑其利润率情况。
If you think that DSV4.1 to V4.1-1017 improvement is solely due to OpenAI distillation, then I wonder why OpenAI couldn't distill a Luna that beats even V4.1. Just how tiny must Luna be then? and how high their margins, given that it's slower too, despite superior hardware? https://t.co/RhIgBFPUxa