Skip to content

Case wall

Jev AI examples, page 22

615 real Jev AI builds and public demos, updated daily. Original text, metrics, and a link back to each source.

Last updated

Public Jev AI examples and demos, not official TypeSafe docs. Clients and SDKs are in the Jev GitHub projects. Paths for jev-latest, Vercel, and OpenRouter are in the September 2026 channel map.

Agent-skill repositories are listed on Jev skills on GitHub.

OpenRouter model ids recorded for Jev AI are on Jev AI on OpenRouter. Rubric and eval posts are on Jev as a judge. How the company and the model fit together is on TypeSafe AI. The pricing calculator uses the published input rate.

Filter

24

Poster from 梭哈.AI (@SUOHA_AI)

Benchmarks

梭哈.AI @SUOHA_AI

ZH

@SUOHA_AI compares Jev and DeepSeek on matching 15 brands to PR hooks in a live news stream and tagging intent. They say Jev finished all 428 items in 28 seconds while DeepSeek V4.1-flash had done 6 of 428. They attribute the gap to Jev returning a decision in one pass instead of writing text, and point to an Elvis Sun GitHub demo.

完全睡不着了....这个视频让你完全看懂JEV的恐怖能力 我实际演示了一下,用 JEV 和 DeepSeek 做同一个任务:从当天的海量实时新闻流里,为 15 个品牌快速匹配有没有适合借势的公关热点,并打上结构化意图分类标签 结果是:28 秒内,Jev 狂刷完了整整 428 条数据,并完成了全部的目标分解与分类;而同一…

Quoted @SUOHA_AI

Poster from Marcus Lowe (@marcus_lowe)

Apps

Marcus Lowe @marcus_lowe

EN

@marcus_lowe built a Tetris demo with early access to TypeSafe’s Jev. They say generation is fast enough that it is pushing the blocks down.

I got early access to @typesafeai's new Jev model and built a demo of it playing tetris The generation speed is so fast that it's pushing blocks down This feels…

Poster from sengpt (@sengpt)

Apps

sengpt @sengpt

TR

@sengpt shows four Jevs playing the Turkish tile game 101 against each other and says the video is not sped up. They treat “make a model play 101” as a personal benchmark because the model must follow the rules and decide from the table state and opponents’ moves.

arkadaşlar jev'e early access aldım ve test etmeye başladım ve gerçekten çok impressed olduğumu söylemem lazım sadece şu kaydettiğim demoyu bi izleyin lütfen. k…

Quoted @CompleteSkeptic

Poster from mikegee (@mikegee)

Apps

mikegee @mikegee

EN

@mikegee controls Figma by voice with TypeSafe Jev. They say the total cost from the first line of code to recording the video was $0.01.

Controlling Figma with voice using Jev from @typesafeai Total cost from first line of code to recording this video: $0.01 🤯 https://t.co/HTGcrBYpTb

Poster from Raghav (@RaghavPunnam)

Guides

Raghav @RaghavPunnam

EN

@RaghavPunnam posts an explainer for people trying to understand how Jev differs from LLMs and how to use it.

For those trying to understand the difference between Jev and LLMs + how to use it https://t.co/55o5WEq3Va

Quoted @CompleteSkeptic

Poster from Steve Krouse (@stevekrouse)

Apps

Steve Krouse @stevekrouse

EN

@stevekrouse says TypeSafe’s Jev is fun and links a live demo at typesafe-demo.val.run. The post does not describe what the demo asks Jev to decide.

typesafe's jev is fun! live demo you can play with: https://t.co/ctnPF0w70r https://t.co/i6kMGkkWOV

Poster from Lahfir (@mdlahfir)

Notes

Lahfir @mdlahfir

DA

@mdlahfir posts “Jev solved delegation” and a link. The text does not say what was delegated or how.

Jev solved delegation https://t.co/1TznArCaSY

Quoted @mdlahfir

Poster from Higgsfield AI 🧩 (@higgsfield_ai)

Agents

Higgsfield AI 🧩 @higgsfield_ai

EN

@higgsfield_ai shows Jev plus Higgsfield routing generation models: Jev reads a prompt and picks models for video and image generation on the Higgsfield API.

Jev + Higgsfield = solved auto-routing for genAI models. In this demo, @typesafeai’s Jev evaluates prompt and picks the most fit models for video and image gene…

Quoted @higgsfield

Poster from 李不凯正在研究 (@libukai)

Benchmarks

李不凯正在研究 @libukai

ZH

@libukai compared Jev on OpenRouter with Gemini Flash Lite on classifying whether People’s Daily articles mention Hubei, using 500 positive and 500 negative examples. They report about 0.35s per article versus 3s, disagreement on about 15% mostly from person names, and about $1 for roughly 5,000 items and 30M tokens.

这两天关于 Jev 这个模型的热度非常之高,甚至可以说是近半年我重点关注了英推以来,看到的原创内容密度最高的一次。 正好我过去一年一直在跑一个小项目,其中有一个环节就是要对每天出版的《人民日报》上的新闻做一个分类,判断每一篇文章中是否包含有湖北相关的元素。 一年跑下来,积累了接近 2 万 4 千条数据,其中判定包含的有…

Poster from kvyb (@0xkvyb)

Notes

kvyb @0xkvyb

EN

@0xkvyb posts “The real use-case for Jev” with a link. The text does not say what that use case is.

The real use-case for Jev https://t.co/fnsghSM7oo

Poster from Shannon 🤖☠️ (@iamMrDuncan)

Training

Shannon 🤖☠️ @iamMrDuncan

EN

@iamMrDuncan built an alternative of the TypeSafe-style API on Cerebras with Qwen 3.8 27B. They say quality and speed were similar, but TypeSafe was much cheaper and beat Qwen on performance. Source is linked.

Built an alternative version of @typesafeai but on @cerebras with Qwen 3.8 27b. Similar quality, similar performance, but vastly different cost. TypeSafe was wa…

Quoted @CompleteSkeptic

Poster from vogel (@ryanvogel)

Apps

vogel @ryanvogel

EN

@ryanvogel built what they call an LLM from first principles with Jev: 29 yes/no questions per character (a–z, space, comma, or period). The highest probability is appended and the updated text is fed back, an autoregressive loop made of a classifier.

i made an llm from first principles with Jev 29 yes/no questions per character: should the next key be a–z, space, comma, or period? highest probability gets ap…

Quoted @CompleteSkeptic

Poster from Cyclotruc (@romdot2)

Agents

Cyclotruc @romdot2

EN

@romdot2 demos pub-sub without a subscriber list: TypeSafe Jev does the routing almost instantly. A demo link is in the post.

pub-sub without sub @typesafeai jev does the routing almost instantly demo link below https://t.co/pTQIUtVBrI

Poster from Ackerman (@Yarilo7brigada)

Apps

Ackerman @Yarilo7brigada

EN

@Yarilo7brigada says they scanned more than 700 live ads with Jev in 40 seconds for 4 cents, and contrasts that with an Opus 5 estimate of 2 to 4.5 million tokens, $15 to $50, and 10 minutes to an hour.

jev is an insanely cool product Saw a post about jev and decided to check it out This thing is straight up gold whoever built it is a genius jev is the future S…

Poster from Vlad Terin (@VladTerin)

Agents

Vlad Terin @VladTerin

EN

@VladTerin built Jev Browser, an open-source adapter for existing Codex browser tools. The video is 1x speed: the user talks, Codex plans, Jev selects, and the browser moves.

I talk. Codex clicks. Watch the browser go⚡ This video is 1× !!! speed. Built Jev Browser: an open-source adapter for your existing Codex browser tools. Say the…

Poster from EDteam (@edteamlat)

Guides

EDteam @edteamlat

ES

@edteamlat explains Jev as a new AI from Diogo Almeida, a ChatGPT co-creator, claiming up to 200x faster and 400x cheaper than current models. Unlike an LLM, it does not generate text word by word: it evaluates options, decides, and returns probabilities.

Jev es una nueva IA creada por Diogo Almeida, uno de los creadores de ChatGPT, que promete ser hasta 200 veces más rápida y 400 veces más barata que los modelos…

Poster from cocktail peanut (@cocktailpeanut)

Notes

cocktail peanut @cocktailpeanut

ES

@cocktailpeanut posts “Silicon Valley predicted TypeSafe” and a clip. The text does not describe a build.

Silicon Valley predicted @typesafeai https://t.co/v8xXPluYF3

Quoted @CompleteSkeptic

Poster from Harshil (@harshil1712)

Apps

Harshil @harshil1712

EN

@harshil1712 demos slides that navigate themselves from speech, built with access to TypeSafe Jev. They apologize for bad audio.

What if your slides are smart enough that they navigate automatically? 👀 Thank you @typesafeai for providing access to Jev! I had this idea for a few weeks, and…

Quoted @harshil1712

Poster from Codez (@0xCodez)

Guides

Codez @0xCodez

EN

@0xCodez quotes Diogo Almeida saying the next era is not the Claude Code era, and that Jev is about 200x faster, 400x cheaper, and, in his words, has zero hallucination. The post says a 17-minute explanation covers a flaw baked in by RLHF.

Jev Founder, Diogo Almeida (ex-OpenAI): "The next era is not the Claude Code era. Claude Code is still part of the assistance era - Jev is the next step Jev is…

Quoted @DeRonin_

Poster from Shane (@digitalshane_)

Apps

Shane @digitalshane_

EN

@digitalshane_ says that with their new Jev access, the model can route website visitors to the pages they need based on what they need help with.

I got access to Jev! It can route website visitors to the pages they need based on what they need help with! https://t.co/AydblKM9Y0

Quoted @CompleteSkeptic

Poster from Niels Rogge (@NielsRogge)

Guides

Niels Rogge @NielsRogge

EN

@NielsRogge introduces a Papers with Code MCP server and quotes what Claude Code returned when asked to research Jev’s architecture with search_papers: a guess that it could be a large schema-conditioned bidirectional encoder with parallel label heads, trained with a proper scoring-rule RL objective. That description is the tool’s output, not a TypeSafe spec.

Introducing the Papers with Code MCP server! To celebrate its release, I've asked Claude Code to research the architecture of Jev by @typesafeai using the `sear…

Poster from Rory Garton-Smith (@rory_builds)

Apps

Rory Garton-Smith @rory_builds

EN

@rory_builds built Easy-Jev, a demo where you change inputs and watch classifications update in real time. They argue for splitting a statement into classifications and routing each to the right kind of model.

I just built Easy-Jev, a Jev demo anyone can try right now! Change the inputs and watch the classifications change in real time. I did my master's thesis on cla…

Quoted @CompleteSkeptic

Poster from Mormon Negro (@mormonnegro)

Agents

Mormon Negro @mormonnegro

ES

@mormonnegro wired TypeSafe and headless Chromium so agents browse faster. In the example the agent starts at the Spanish Wikipedia page for coffee and follows links to artificial intelligence in 20 seconds.

Acabo de terminar la implementación de @typesafeai + Chromium Headless para que mis agentes puedan navegar por internet a una buena velocidad! En este ejemplo l…

Quoted @mormonnegro

Poster from Ashutosh Mathore (@ashutoshftw)

Benchmarks

Ashutosh Mathore @ashutoshftw

EN

@ashutoshftw had Jev, Claude Haiku 4.5, and Gemini 3.5 Flash-Lite play Tetris with the same 200 pieces and the same legal moves. They report Jev at 9200 points, 75 lines, 300ms per move, and 0 errors, ahead of Claude (8900, 1.52s, $0.48) and Gemini (9000, 1.13s, $0.02).

Had jev, claude haiku 4.5, and gemini 3.5 flash-lite play tetris. Same 200 pieces, same legal moves, real time. Jev: 9200 pts, 75 lines, 300ms/move, 0 errors Cl…

Unofficial developer resource. Not affiliated with TypeSafe AI. These posts belong to their authors. jevnotes.com does not host the videos.