一个只会做选择题的小模型，能长出多少玩法？ TypeSafe 的 Jev 就是这么个东西：给它最多 255 个选项，几百毫秒挑一个，再附一个校准过的置信度，输出 token 免费 结果开发者们已经拿它做出了 19 个开源项目，加起来 6800 多星 挑几个最有意思的 jev-ultrafast，browser-use 出品，2700 星，一句 Show more
Core Concepts
System Two task
Work that needs extended, multi-step reasoning; the docs treat it as a signal to decompose rather than a job for Jev.
- Category
- Core Concepts
- Also known as
- —
- Related terms
- 4
- Directory entries
- 2
- Docs
- docs.typesafe.ai
- Added
- 2026-09-24
Definition
The jaggedness page lists System Two tasks — more layers of indirection, extended reasoning — among the things to avoid. Jev is built for fast, focused judgments, and its accuracy falls as a question requires multiple hops.
The remedy is not a longer prompt but a different decomposition: split the reasoning into judgments code can sequence, or hand the genuinely deliberative step to a generative model.
Tagslimitationsdesign
Related terms
Definitions that connect to this one.
Atomic questionA question that asks one specific, well-scoped thing — the kind of judgment a knowledgeable person makes in a few seconds.Core ConceptsIndirectionQuestions about a property of a property, or with double negatives, cost accuracy; write instructions as directly as possible.Evaluation & TrainingJaggednessThe documented set of jev-1.13 failure modes — literal reading, counting, dates, indirection, context rot — each with a guardrail.Evaluation & TrainingAgent loopA control pattern where the model chooses its next step repeatedly; TypeSafe's guidance is to prefer a code-owned workflow instead.Core Concepts
From the directory
A maintained list of jev-1.13's known failure modes, literal reading, unreliable counting, dates as text, indirection, context rot, and contradictory criteria, each with a guardrail.
Practices & PatternsDocs#official#docs#evaluation
Official
Design AI-powered software by keeping code in control and giving System One narrow, structured decisions to make.
Practices & PatternsDocs#official#patterns#architecture
Official
From the community
Posts from builders shipping with Jev right now.
19 open-source Jev projects
Jev Calc: a notebook that calculates anything
i made a smart calculator notebook using jev it can calculate ANYTHING!! 😅
Day-one notes from a skeptic, in Chinese
一些关于 Jev (@typesafeai) 的 notes 研究了一天 Jev，带来的新鲜感迅速回落，这好像就是一个更快的通用分类器/决策器，LLM 完全可以做到。 而且因为不了解模型背后的参数规模（应该不会很大），世界知识可能不一定有常规 LLM Show more
@CompleteSkeptic
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x
Search and tagging on keep.md
Cloudflare Workers has Jev now so I'm putting it to the test on keep.md - 7x faster search rerank compared to the current hybrid - 50x faster tagging of content vs GLM 4.7 Flash with no failures
A filter for reply-guy comments
Pro tip: You can use Jev to remove annoying reply guy comments that X seems to always miss. Takes 5 minutes in Astra with the docs and an API key.
A Downloads folder that sorts itself
Jev unlocks SO many awesome new ideas. I built a macOS app that monitors my Downloads folder along with a customisable set of rules. Is the downloaded file an invoice? Move it to a special folder with the correct filename. No other LLM calls involved - just Jev!