Allie the Icon
@allietheicon
One of my favorite categories of Jev use-cases is real time speech! Teleprompters, speech coaches, and so much more
XTools and apps
Updated Nov 13, 2026
855 projects, posts and guides about Jev, the System One model from TypeSafe AI. Each entry links to its source and shows the cost and speed its author reported.

Diogo Almeida
@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 cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution
Allie the Icon
@allietheicon
One of my favorite categories of Jev use-cases is real time speech! Teleprompters, speech coaches, and so much more
XTools and apps
Erik Spock Gafni (hiring!)
@EGafni
my personal faves are: AutoResearch for Feature Extraction Use LLMs to generate Jev questions's who's probabilities are fed into a classic ML algorithm like logistic regression to predict a label. Hierarchical Classification Traverse a classification taxonomy using Jev and beam sesarch. Jev is great for graph traversal.
Guidex.com
Erik Gafni of TypeSafe names his favourites from the cookbooks. AutoResearch for feature extraction: an LLM writes Jev questions, and their probabilities feed a classic algorithm such as logistic regression to predict a label. Hierarchical classification: traverse a taxonomy with Jev.
Dorian Smiley
@dsmiley411
I finally managed to burn through my free usage credits with Jev, all $5 of them. 123M tokens. Man this thing is cheap!
Tatsuya Shirakawa
@s_tat1204
jevや派生を使ったアプリケーションも、あまりconfidenceを利用していなさそうなことを考えると、識別ができてしまえばなんでも良く、この手のembeddingベースの方法でも十分なケースは多そうですね。
XTools and apps
Aradhye Agarwal
@AradhyeAgarwal
While the whole world is going crazy over Jev, I give you a more skeptical take. I think it's a great model––but its modeling choices cause real failure modes. 1. The way the speed argument is presented in the quoted tweet isn't valid IMO. You can simply treat every row in the output as independent and sample the autoregressive LLM in parallel. 2. The main benefit of autogressive models is enabling increased computation depth with the same amount of model parameters (via CoT). I don't see how that would be possible with Jev. 3. Jev isn't a magic wand that will one-shot any decision task. I can come up with a task where LLMs do better in terms of efficiency: suppose you have 100 choices out of which you have to choose 10. Also suppose someone has already chosen some 10 choices. You have to tell the "goodness" of this choice. With Jev you could naively multiply the probabilities of choosing the options, but this would not work. Making one choice changes the probability distribution of the other options which has to be accounted for. A simple example one of the chosen options is repeated. Jev will give both the options equal probability (due to symmetry) but it's unlikely that this would be the optimal choice (assume repeating options is downright bad). With LLMs however you can prompt it like so: "okay you have 100 options and you have to answer which ones (a total of 10) you would choose in a sorted order." Then gaslight the LLM into thinking it generated the 10 options (in sorted order) by just prefixing those tokens to the assistant channel and then measure the probability of generating that output. This would happen in one-shot. A careful reader might say: "but we can do the same thing with Jev too right?" Yes that's correct, you could have Jev choose one option then conditioned on that choose the second one and so on. Then multiple the probabilities. But the problem is that this will take 10 sequential steps; you can't one-shot it unlike with an LLM where there is no recurrence.
XSales and leads
Trackmind
@0xTrackmind
x.com/i/article/2106…
Guidex.com
Trackmind’s X article on Claude loops that run unattended, with Jev as the decision layer.
Jun Penn
@heyjunpenn
Hey Jev builders! 🚀 What are you working on? Get your demo into JevBest so people can watch it and find the original post. Drop yours below 👇
XTools and apps
Sergii Makarevych
@sermakarevich
@typesafeai JEV use case I will have first AI coding interview ever. I know its gonna last for 1.5 hours and I will have to clone an exercise repo. So what I should be prepared for ? Lets use LLM for hypothesis and JEV for ranking:
XTools and apps
Jerry Xu
@jerrycxu
tab-jev: jev-like model + tabular foundation model = an in-context learner for your text & tabular data. https://github.com/edamame-labs/tab-jev Many real industry datasets are a mix of tabular and text data: - Tabular foundation models like TabPFN learn from a few hundred labeled rows in context, with no training. But they can't read text. - Jev-style models read text and answer typed questions with scores. But they can't digest tabular data well. Experiment results on kaggle problem "Kickstarter: will the project get funded?" is shared in the repo. Missing a flight at JFK is a productivity booster...
XContext and memory
Asteri
@Asteri_eth
Claude + Jev is the first agent setup where I stopped babysitting every single step most agents burn frontier tokens on tiny calls like "retry?" or "enough sources?". this one hands them to a model that does 10,000 decisions for ~$0.42 10 min and it runs on your machine: prompt → Claude → Jev picks the move → gate checks confidence → Claude executes → logged result step 1 → get your Jev API key on the TypeSafe site step 2 → clone the router from the GitHub below and install it in one line: python3 -m venv .venv && .venv/bin/pip install -r requirements.txt && cp config.example.yaml config.yaml step 3 → export the key before the first run: export TYPESAFE_API_KEY='YOUR_KEY' put the same line in ~/.zshrc or ~/.bashrc so a fresh terminal doesn't lose it step 4 → give Claude skill/jev-decision-layer.SKILL.md and point it at src/router.py + recipes/. on its own Jev only returns probabilities, the router is what turns them into actions Claude can run step 5 → test the full chain end to end, not just the Jev endpoint: .venv/bin/python -m src.cli '{"goal":"what is 2+2?","kind":"chat"}' the JSON you get back should have action, reason, mode, jev_used and confidence step 6 → start in mode: shadow. Claude works like before while every Jev call is logged next to what Claude actually did. after 20-50 real decisions keep only the question packs that matched step 7 → flip to mode: active with hard gates: 0.80+ acts on its own, 0.50-0.79 only suggests, under 0.50 comes back to you what changed for me: Claude stopped wasting tokens deciding whether to retry, reuse the cache or spin up a subagent. Jev makes those calls, Claude does the actual work, and anything it isn't sure about lands on my desk instead of in my outbox grab the exact stack from the repo, then read how every gate works in the full Jev architecture breakdown ↓
Guidex.com
Asteri’s setup hands the small calls an agent makes, such as “retry?” or “enough sources?”, to Jev and not to a frontier model. The post gives about $0.42 for 10,000 decisions and says it runs on your machine.
Eugene Cheah - AI builder @ 🇸🇬|🇺🇸
@picocreator
Yes, it is quite literally next to the score - the preferred prompt format - and so is the code for the inference at github.com/featherless-ai…
XTools and apps
belorix
@0xbelorix
YOUR VERIFIER STILL COSTS LIKE A FULL MODEL Claude writes. then Claude reads the whole pile again to say yes or no. every extra loop turn makes that pile bigger. Jev: $0.042 per 1M input tokens. output tokens free. it returns a decision, not another essay. same job. wrong tool if you still use a writer as the gate. read the article below
Stephon Proctor, PhD
@stephonomon
Deskilling is a real risk when it comes to using AI scribes. The drafts often look plausible, which leads users to overly trust without verifying. Models like #Jev could help provide users with better insights onto areas where the draft might be wrong. github.com/Stephonomon/sc…
XTools and apps
shawn
@shawnbuilds
Claude Mods are like custom Minecraft servers Claude Code Here's how to use Claude Code Mods + Jev to build your own custom memory harness for Claude Code. It runs in milliseconds, at sub-penny cost compared to subagents ⬇️ x.com/i/article/2106…
Guidex.com
shawn’s step-by-step guide: Claude Code Mods plus Jev make a custom memory harness for Claude Code. He says it runs in milliseconds at sub-penny cost compared to subagents.
Likit D
@Likitd_
Jev jev jev jev Your RAG pipeline just got a relevance upgrade. excited to introduce jev-ranker — Jev-powered reranking and relevance filtering How much can smarter reranking improve a RAG pipeline? Try : npmjs.com/package/jev-ra… Benchmark Results:
XSearch
Lahiri
@Ziggyzonty
JEV is INSANE... I gave it an educational video and its transcript. Every 5 seconds, it looks at the next 5 seconds of content in the context of the previous 60 seconds. It scores the segment on 5 dimensions: Relevance. Usefulness. Importance. Memorability. Actionability.
XTools and apps
James Pardoe
@JamesPardoe
If you're still scratching your head about Jev, this is the best explanation that I've seen.
XBenchmarks and evals
Ethan Walker
@ethan_walker_ai
🧵 What if the next big AI model isn’t designed to chat? TypeSafe AI is pushing a different direction with Jev, a model built to make structured decisions for software. No essays. No long answers. Just decisions that software can act on. Here’s why this matters 👇
Guidex.com
Ethan Walker’s explainer thread: Jev is built to make structured decisions for software. No essays and no long answers, only decisions that software can act on.
Dorian Smiley
@dsmiley411
Testing Jev’s accuracy tonight for next best action prediction. The results: Canonical accuracy: 98.6% Generalization accuracy: 41.5% We ask Jev to predict the next state in a program from the current partial program. The suite has 25 cases and we ran it 20 times. Seven cases are represented in the in context examples. The other 18 are held out cases that require Jev to generalize from those examples. The failures are not uniformly random. Jev generalizes perfectly on some unseen compositions and fails almost deterministically on others. That suggests there may be specific structural boundaries to what it can infer from context. Maybe some of this is prompt design. Maybe it is a capability boundary. We’re testing that now. But next best action prediction is important. A huge amount of software today contains really brittle decision logic: onboarding, payments, claims, revenue cycle, approvals, exception handling, etc. If Jev is a bet on software consuming intelligence, this is exactly the kind of high frequency, high value logic it needs to improve.
XBenchmarks and evals
Jason Alco
@Jasonalco
I put Jev vs 5.6 Sol head-to-head on a text categorization step in a platform I’m building. Surprised by the outcome. Aren’t text-based judgements the exact use case Jev is made for?
XBenchmarks and evals
Mendy
@MilkMendy
If you run a business, Jev might be a really good alternative to your heavier models. It's really good at classifying things and choosing between yes or no, and it could be 200X quicker and a lot cheaper than the other frontier models. #AI #AIAgents #AIAutomation
XBenchmarks and evals
Guidevercel.com
Vercel’s six-minute guide. Ask two narrow questions about one diff: which review track fits the change, and does it require existing callers to migrate. Application code collects the evidence and uses the answers to organise review. The method carries over to any decision model.
Sama
@osamalamaaa
all that hype about the potential use cases for jev and everyone just rebuilds common programs from first principal. I have yet to see a legit new use case that has wowed me but thanks random guy for building chatwithjev real novel stuff
XTools and apps
@kleos
@1kleos1
A 20-YEAR-OLD JAPANESE STUDENT BUILT A JEV POLYMARKET BOT OVER THE WEEKEND, FIRST NIGHT IT MADE $17,831 Starting capital: $148 First night: $17,831 profit Total profit so far: $276,000 he built the whole thing around one stupidly simple loop Polymarket data comes in → Jev judges the setup → confidence gate checks it → deterministic code executes Jev only gets three choices BUY YES BUY NO WAIT the bot scans dozens of markets at once looking for odds that stop matching the live data around them while a normal LLM is still explaining why something might be mispriced, Jev has already made the decision and moved to the next market position sizing, max loss, execution and exits stay in code Jev only handles the fuzzy part is this market actually wrong right now? $148 starting capital $17,831 on night one $276,000 so far and the craziest part is I have the full setup he used the Jev prompts, decision flow, confidence gates and Polymarket architecture might drop the whole thing next
Sodeh Abadi
@irSodeh
Jev vs Laya-MLX isn't really a rivalry. Jev is the original, commercial model. Laya-MLX is a local, open-source take on the same idea, running natively on your Mac. Same core concept: "System One" models. No chatting, no text generation. Just fast, typed decisions. #Jev #Laya
XOpen source
Guidetechcrunch.com
TechCrunch on the round: TypeSafe AI raised $870 million at a $7.5 billion valuation, led by Andreessen Horowitz with Sequoia and DCVC. The company claims a third of the Fortune 500 already use Jev.
WquGuru
@wquguru
Jev 发布才 8 天,开源替身已经一堆了,今天诺基亚也来蹭热闹了😅 我在 MacBook 上,让官方 Jev 和 3 个开源方案同跑 20 张中文工单🫱 Laya 25ms,比 Jev 快 50 倍 djev 509ms 官方 Jev 1.3 秒 AnyJev 将近 4 秒 仓库都在这: Laya(Mac 版)http://github.com/mizorewww/laya-mlx djev(Mac 版)http://github.com/jamescorbett/mlx-vlm AnyJev http://github.com/nokia-applied-research/AnyJev 但快的,不一定对 👇
Darshan Jain
@i_darshanjain
Jev is all over my timeline rn 👀 Went looking into it and stumbled across Laya. Open-source System 1 decision model. They even have a Jev comparison 👀 7.8× faster. 3× better calibrated. 45/51 languages. Self-hosted. And the weights are right there. 😂 Yeah… I am watching this space. https://huggingface.co/convaiinnovations/laya
XOpen source
Saksham Malhotra
@SakshamMalhot27
Launching JevRelevanceRetriever just $0.73 across 6,460 calls Open src Jev LangChain BaseRetriever for rel scoring Install and use it, no chain changes ;) pip install jev-relevance Repo-https://github.com/saksham-malhotra-27/jev-relevance Open for issues and PRs, feel free to contribute
XSearch
PickGuidea16z.com
Andreessen Horowitz explains why it led the round. It calls the Jev launch the biggest narrative violation it has seen this year, and says Jev reached 1 trillion tokens generated in 3 days after launch.
Koimiao🐈
@jaunatis_q
We built a Stanford Town powered by Jev. It isn’t a game demo, there is no player. The residents decide where to go, who to meet, and when to talk. Humans can only watch their pixel world unfold. Demo + code: github.com/NevaMind-AI/je… #Jev #AIAgents
XGames and real time
PickNitin.nn
@NitinthisSide_
After seeing Jev everywhere on the feed lately, we had to try this direction ourselves. We’re bringing Laya to Insight O Mate. A fast, open-source, local decision model that fits perfectly with our privacy-first approach to querying NoSQL databases. Your data stays local. Your decisions stay local. Insight O Mate + Laya is coming. Try Insight O Mate at https://insightomate.lunamic.co
XOpen source
Choroshin Alex
@choroshin
"לקוח יקר, זוהתה פעילות חריגה בחשבון הבנק שלך. החשבון יוקפא תוך 24 שעות. לאימות פרטים לחץ כאן." רציתי לשחק עם Jev, המודל החדש של TypeSafe, וגם לבנות איתו משהו שיש בו תועלת אמיתית. אז בניתי ב-Skills IL כלי שבודק הודעות כאלה: מדביקים את ההודעה, מ-SMS, מוואטסאפ או ממייל, ומקבלים
XSecurity and abuse