Skip to content
learnjev

TypeSafe AI shipped Jev on 15 September 2026

The model that won’t talk to you — and why that’s the point.

Jev is a System One model: unstructured state in, typed probabilistic decisions out. No strings, no parsing, no retry loop around json.loads. These are independent, hands-on tutorials for it — written the week it launched, with every claim traced to a source.

One request

POST /v1/systemone
{
  "state": "Our API started returning 500s 20 minutes ago
            and we can't process orders until it's fixed.",
  "model": "jev-latest",
  "questions": {
    "department": {
      "type": "choice",
      "instructions": "Which team should handle this",
      "criteria": {
        "billing": "Payment or subscription issues",
        "technical": "Bugs or integration problems",
        "sales": "Pricing or account questions"
      }
    },
    "frustration": {
      "type": "score",
      "instructions": "How frustrated the customer appears",
      "criteria": ["Calm", "Frustrated but civil", "Very angry"]
    },
    "is_urgent": {
      "type": "noul",
      "instructions": "The message conveys urgency"
    }
  }
}

Three typed answers, evaluated in parallel

departmentchoicetechnicalconfidence 0.82
  • technical
    0.85
  • billing
    0.08
  • sales
    0.07
frustrationscore1.60confidence 0.78
  • Very angry
    0.65
  • Frustrated but civil
    0.30
  • Calm
    0.05
is_urgentnoul0.999

no ← 0.5 → yes · nouls carry no confidence field

No prose. No JSON inside a string. Nothing to parse, nothing to validate, and no code path for “the model replied with an apology instead”.

$0.042/ MTok in

output not charged

70–500ms

TypeSafe's figure

255options

max Choice cardinality

64ktokens

32k state + longest question

System Two vs System One

It gives up strings. In exchange it gets four things back.

Three of them are speed and cost arguments. One of them — calibration — is the thing JSON mode on your existing model genuinely cannot give you.

Outputs
An LLMStrings — a chat response, code, a refusal, or a hallucination. Must be parsed and validated.
JevType-safe values from a set you defined in advance. Never a type error.
Sampling
An LLMSequential. One token at a time, each conditioned on the last.
JevParallel. Every question in a request answered in one shot.
Confidence
An LLMOverconfident and inconsistent, even when you ask for a number.
JevA calibrated probability with every Choice and Score.
Explains itself
An LLMYes — that is most of what it is for.
JevNo. There is no rationale field, and none is promised.

Condensed from TypeSafe’s launch post. The last row is ours — it is the constraint most likely to rule Jev out for you.

Before you commit

Three things the launch coverage got wrong

We like this product. We also read the footnotes. Every number below is TypeSafe’s own — the corrections come from their documentation, not from a competitor.

  1. 01

    “Never hallucinates”

    It never makes a type error, which is structural and genuinely useful. It can absolutely return the wrong valid answer. The 0% bar on TypeSafe's own chart is, in their words, “not empirical” — it is added by definition.

    Read the full argument →
  2. 02

    “193.6× faster, 444.6× cheaper”

    One best-case demo, which TypeSafe itself flags as “on the higher end”. Matched on accuracy against its nearest peer it is roughly 76× cheaper and 25× faster — and 6.3 points behind the best baseline. Independent testers measured 3–11×, and on one many-class task Jev simply lost.

    Read the full argument →
  3. 03

    “Calibrated probabilities”

    TypeSafe published no ECE, Brier score or reliability diagram at all. A pre-registered community study since has: calibrated on one dataset (ECE 0.020), overconfident on another (0.094). Calibration is dataset-dependent, so the only number that matters is the one from your data.

    Read the full argument →

Latest

It has already changed since launch

Full timeline

Tutorials

Start at the first call. Finish in production.

All 13 tutorials

One request. A dozen typed decisions. No parsing.

Whether or not Jev is the model you end up shipping, the architecture it forces on you — decomposed judgments, thresholds in code, policy in version control — is worth adopting today.