X · by Kenny
2,300 AI papers in about 83 seconds
A small-team case, with a caveat about where the cost came from.
Kenny Chen｜AI 实战
@KennyChinaTech
Jev 这个案例很适合小团队：2,300 篇 AI 论文，约 83 秒，成本 $0.14。 但这个数字不能直接当成 Jev 的单模型成本。旧标签先由 DeepSeek V4 Flash 跑过，真正该测的是整条分类链路：预处理、Jev 决策和人工抽检加起来还剩多少。 x.com/omarsar0/statu…
Sep 22, 2026 · 0 likesOpen on X
The author reads the case as a fit for small teams: 2,300 AI papers classified in about 83 seconds for about $0.14. They caution that this is not the cost of Jev alone, because the older labels were produced earlier under different conditions.
Open the source
