Skip to content

explainer · updated 2026-09-21

What is Jev? The System One model, explained

Jev is an AI model from TypeSafe AI that decides instead of writing. Give it any text or JSON and a few typed questions: it answers all of them at once, with a probability for every possible answer. It is the first System One model.

maker
TypeSafe AI
launched
2026-09-15
answers
choice · score · noul
price
$0.042 / M input tokens
median latency
0.44–0.48 s
model
jev-1.13.0

What is a System One model?

An AI model that answers typed questions with probabilities, all in one pass, and never writes text.

System 1 · fast

One look, one answer

Jev: every question answered at once, as probabilities.

System 2 · slow

Step by step, in words

LLMs: text one token at a time, reasoning by writing.

The name borrows the fast, intuitive System 1 of Daniel Kahneman's Thinking, Fast and Slow. Most decisions software makes are that kind: spam or not, which team, safe to run.

typed

Answers only with your options. Nothing to parse.

parallel

Every question in one pass. Ten take as long as one.

probabilistic

Odds for every answer, not just a pick.

cheap

No text is generated, so output is free.

How does Jev work?

State in, decisions out. One API request carries the input and every question, and each question is answered on its own.

STATEJEV · EVERY QUESTION IN ONE PASSYOUR CODE{ "ticket": {"from": "Priya N. (Pro plan)","subject": "Payouts failing again","text": "Third payout this week bounced with error PAYOUT_REJECTED. I have to pay two contractors by Friday and your status page says everything is fine. What is going on?" }CHOICEWhich team should handle this ticket?billing 0.97NOULDoes the customer need a reply today?P(yes) 0.89SCOREHow frustrated is the customer?expected level 2.34 of 3switch (team) { case "billing": queue(ticket)}if (today > 0.9) replyFirst()tickets.sort( by(mood))ACT
The ticket above, taken apart. Each question runs in its own lane.
state
What the decision is about: any text or JSON. The context window is about 32,000 tokens, text only.
question
A typed ask, choice, score or noul, with an instruction.
criteria
The possible answers, described in plain words.
independence
Questions never see each other's answers. Combine them in code.

What are choice, score and noul?

Jev's three question types: pick one option, place on an ordered scale, or answer yes or no.

choice match

Pick one of up to 255 options.

request state + question

message
Can I move my table from Friday to Saturday night?

What does the customer want?

choicebookchangecancelaskwith descriptions

for
routing, intent, next action
fields
choice, probabilities
note
Probabilities sum to 1, so add an other option.

score sort

Place it on 2 to 10 ordered levels.

request state + question

review
Food was fine and the service slow, but the view made up for it.

How positive is the review?

scoreNegativeMixedPositiveGlowing

for
severity, quality, ranking
fields
score, probabilities
note
The score is the expected level: sort by it.

noul if

Yes or no, as the probability of yes.

request state + question

email
Your parcel is held at customs. Pay the $2.99 release fee within 24 hours: parcel-release.help/pay

Is this a phishing attempt?

noulyesno

for
flags, checks, gates
fields
noul
note
Short for Bernoulli; boolean in the Vercel AI SDK.

How is Jev different from ChatGPT and other LLMs?

An LLM writes its answer token by token, and your code parses it. Jev is not an LLM: it writes nothing and scores your answers, all at once.

LLM · one token at a time

Jev · every answer at once

LLMsJev
OutputText to parseProbabilities over your answers
Writes text or codeYesNo
Reasons step by stepYesNo: one look per question
Latency (median)0.53–2.21 s0.44–0.48 s
Cost per 1,000 decisions$0.023–$1.5$0.029–$0.043
Possible answersAnything, including made-up onesOnly the options you defined
ProbabilitiesOnly if asked to write themBuilt in, trained to be calibrated
Best atOpen-ended workMany fast decisions over known answers

Measured on the Jevals boards: Jev and 6 LLMs, release 2026-09-18.

Is Jev just a classifier?

Close: it does zero-shot classification, and critics say so. The difference from a trained classifier: you describe the options in words and change them on any request.

Trained classifierLLM with structured outputsJev
Training dataNeededNoneNone
New answersRetrainEdit the promptEdit the request
OutputScores per labelText that should parseProbabilities per answer
Speed and costFastestSlowestFast

What do Jev's probabilities mean?

They are meant to be calibrated: answers given at 80% should be right 80% of the time. That lets code act alone when Jev is sure, and ask for help when it isn't.

sure: act

unsure: ask

Same pick, different confidence. Illustration.

yes/no

10said 55%, right 58% (48 answers)said 65%, right 73% (109 answers)said 75%, right 82% (170 answers)said 86%, right 91% (371 answers)said 94%, right 98% (802 answers)

gap 5.0 pts

pick-one

10said 29%, right 0% (1 answers)said 35%, right 20% (25 answers)said 44%, right 23% (40 answers)said 54%, right 51% (69 answers)said 64%, right 54% (69 answers)said 74%, right 63% (95 answers)said 85%, right 60% (131 answers)said 98%, right 91% (1070 answers)

gap 9.8 pts

rubric scores

10said 27%, right 47% (34 answers)said 36%, right 31% (118 answers)said 45%, right 31% (261 answers)said 54%, right 31% (339 answers)said 65%, right 41% (324 answers)said 75%, right 53% (222 answers)said 83%, right 61% (157 answers)said 94%, right 76% (45 answers)

gap 19.7 pts

Jev's real answers, release 2026-09-18: stated confidence across, accuracy up, perfect on the diagonal. Asked for probabilities, the best LLM on each board comes closer (Gemini 3.8 Flash 2.0, DeepSeek V4.1 Flash 2.8, GLM-5.3 12.9).

What is Jev used for?

Decisions with known answers that software makes often: route, check, rank, choose.

  1. 1 · propose

    Code or an LLM lists the options.

  2. 2 · decide

    Jev scores them against the state.

  3. 3 · gate

    Code checks the confidence.

  4. 4 · act

    Sure: do it. Unsure: escalate.

choice

Routing

Send each ticket, email or prompt to the right place.

noul

Guardrails

Approve an agent's command, or ask a human.

choice · noul

Moderation

Publish, redact or reject.

score

Judges and linters

Grade outputs against a rubric, in place of an LLM-as-a-judge.

score

Ranking and reranking

Score every search result, then sort.

choice

Classification

Label a million rows by topic, intent or sentiment.

choice

Agents

Pick the next action or the element to click.

choice

Real-time loops

Game bots deciding several times a second.

choice

Extraction

Code finds candidates, Jev picks one or none.

Limitations: when not to use it

  • Writing anything: replies, summaries, code.
  • Answers you can't list in advance.
  • Arithmetic or multi-step reasoning.
  • Security boundaries: text in the state can sway it.
  • Its documented weak spots: counting, dates, negation.

How does Jev score on benchmarks?

Tied for first on yes/no and tied for second on pick-one, for a fraction of most LLMs' cost. On rubric scores, no model clearly beats guessing.

BoardTestJev Decision ScoreJev accuracyBest other model
noulPubMedQA69.091.3%Gemini 3.8 Flash 73.0
choiceBanking7767.879.7%Gemini 3.8 Flash 74.1
scoreHelpSteer2 helpfulness9.241.3%GLM-5.3 7.8

Scored against human labels, five runs per question; orange marks each board's top 2. Head to head: Jev vs Gemini 3.8 Flash.

Who makes Jev?

TypeSafe AI, a San Francisco startup led by CEO Diogo Almeida, a former OpenAI researcher on InstructGPT.

  1. TypeSafe AI founded, in stealth
  2. Talks on what comes after RLHF
  3. Jev launches
  4. On Vercel AI Gateway
  5. First independent benchmark
training
RLCD, Reinforcement Learning for Calibrated Decisions. Unpublished, like the architecture.
name
After the Jevons paradox: make something cheaper and use grows.
access
TypeSafe's API (waitlist), or Vercel AI Gateway as typesafe-ai/jev.
funding
$40M seed round led by DCVC.

What is a Jev-type model?

Any model that answers choice, score and noul with probabilities. Jev itself is closed; adapters and open-source alternatives now copy its interface.

ModelWhat it isOpenOn Jevals
JevThe native System One modelNoListed
LLMs via an adapterAn LLM prompted for the same probabilities (system-one-adapter, AI SDK evaluate)Varies6 listed
openjevOption probabilities read from an open Qwen modelMITComing
decider-2bA 2B Qwen model fine-tuned for the three typesApache-2.0Coming
DiffusionGemmaGoogle's diffusion model filling an answer template (open-jev)Open weightsNot yet
GLiClass, GLiNERSmall zero-shot encoder classifiers, the prior artApache-2.0GLiClass coming

What are the criticisms of Jev?

Mostly fair: it is close to a zero-shot classifier, and most claims about it are the vendor's own.

  • "Just a zero-shot classifier." Partly true: encoders like GLiClass came first. New is one hosted model, three types, calibration as a goal.
  • "Can't hallucinate." It can't invent an answer, but it can pick the wrong one, confidently.
  • Vendor benchmarks. TypeSafe's Workflow Evals grade agreement with two LLMs, not human labels.
  • A black box. No paper, no parameter count, no published calibration.
  • The price. Critics suspect a subsidy; TypeSafe says it is profitable.

Facts checked 2026-09-21 against TypeSafe's and Vercel's public pages. Jevals is independent of TypeSafe AI.