10 个用 Jev 做结构化决策的开源项目清单
作者追踪一次客服运行发现 44 次模型调用里 37 次只用来选下一个分支，于是列出 Jev、laya、jev-mcp 等专做类型化选择的仓库。
THEY RAISED $40M FOR AN AI THAT CAN’T WRITE, CAN’T CODE AND CAN’T EXPLAIN ITSELF in one traced support run, 37 of 44 model calls only picked the next branch, and these repos are built for that part 01 langextract (37.3k ⭐) structured information pulled out of text, with source grounding ▸ https://github.com/google/langextract 02 dspy (36.7k ⭐) programming language models instead of prompting them ▸ https://github.com/stanfordnlp/dspy 03 laya (30.3k ⭐) an open decision engine with typed choice, score and yes/no answers ▸ https://github.com/NandhaKishorM/laya 04 guidance (21.8k ⭐) a language for controlling large language models ▸ https://github.com/guidance-ai/guidance 05 jev-ultrafast (20.9k ⭐) a web agent where jev picks every browser action ▸ https://github.com/browser-use/jev-ultrafast 06 pydantic-ai (19.6k ⭐) agents that are typed end to end ▸ https://github.com/pydantic/pydantic-ai 07 outlines (15.9k ⭐) structured outputs from any model ▸ https://github.com/dottxt-ai/outlines 08 fast-jev-compaction (7.2k ⭐) a claude code plugin that scores every tool call with jev before compaction ▸ https://github.com/tamaratran/fast-jev-compaction 09 jev-review (650 ⭐) a staged code review workflow built with jev ▸ https://github.com/devagrawal09/jev-review 10 jev-mcp (507 ⭐) typed judgments from jev as mcp tools ▸ https://github.com/jkudish/jev-mcp a model that only returns typed answers can be wrong, but it can’t be malformed
这条是引用着另一条写的：x.com/thegreatest_sv/status/2102802378378621355