jevPOST
Decision models for evals: Jev, Decider, Clef and more
A roundup of probability-output decision models like TypeSafe Jev, Perplexity Decider, Cloudflare Clef and OpenAI Decisions API, plus a recommended evals read.
Dongxi 东锡 NLP
@dongxi_nlpArchived 8 days agoOriginally posted 2026.10.21
Dongxi 东锡 NLP @dongxi_nlpOriginally posted 2026.10.21
好文！ 近段时间涌现出“决策模型”，如： TypeSafe 的 Jev Perplexity 的 Decider Cloudflare 的 Clef OpenAI 的 Decisions API 输出概率而非长文本，常被用作 LLM-as-a-Judge 的替代方案。 如何用这些决策模型做 Evals，推荐读这篇！
This post quotes another post: x.com/hamelhusain/status/2112588257104216294
x.com/dongxi_nlp/status/2112652766997103016Archived 2026.10.21 10:03