CrossModel
Jev
typesafe/jev
Back to model catalog

TypeSafe · Model guide

Jev

Compare
typesafe/jevReleased Sep 19, 2026

TypeSafe's System One decision model. It answers a set of typed questions about a shared input: yes/no predicates, single-choice and graded scores, each with a probability, in one call. Called through /v1/systemone or /v1/decisions, not the chat endpoints. Text input only. The input plus the longest question may take up to 32K tokens; a whole request, all questions included, up to 64K.

Modalities
TextDecision
Context
32K
Max output
—
Price / 1M tokens
$ 0.042input only
Decision modelYes / noSingle choiceGraded scoreImage input

Returns probabilities, not text. Call it through /v1/decisions or /v1/systemone.

What this model is good at

What the vendor positions it for, and which of our models to reach for instead.

Best for

  • Classification and routing at volume · Each call answers a set of yes/no, single-choice and graded questions with calibrated probabilities instead of generated text.

Neighbouring models
Compare them side by side →

Pricing and billing

Billed on input tokens only. There is no output or cache price.

Input sizeInput / 1M tokens
Any size$ 0.042

The whole input counts, including question text and option descriptions. Output tokens, where the upstream reports them, are not charged.

What it answers

A decision model takes one shared input and a set of typed questions, and returns a probability for each.

Context window
32.0K tokens
Max output
—
Input / output modalities
Text → Decision
Question types
Yes / no · Single choice · Graded score
Output
Probabilities per question — no generated text
Streaming
Not applicable — one response per call
Image input
Not supported
Endpoints
/v1/decisions · /v1/systemone
First-party sources:Specsdocs.typesafe.ai 1docs.typesafe.ai 2Capabilitiesdocs.typesafe.ai 1docs.typesafe.ai 2

Frequently asked questions

What is Jev?
Jev is a decision model, available on CrossModel as typesafe/jev. Instead of generating text it answers a set of typed questions about one shared input — yes/no, single choice, graded score — with a probability for each. Its context window is 32K tokens.
How much does Jev cost?
Currently $ 0.042 per 1M input tokens. Decision models are billed on input only — there is no output price.
What kinds of questions can Jev answer?
Three: yes/no predicates (a probability), single choice (a probability per option) and graded scores (a probability per level). It takes text input only.
Which endpoint do I call?
Call POST /v1/decisions (the OpenAI decisions shape) or POST /v1/systemone (the System One shape). Either works for every decision model. The chat endpoints do not accept decision models.
Can I try it without writing code?
Yes. Sign in and open the Playground on this page: fill in an input and a few questions and see the probabilities — no API key needed. Usage is billed from your wallet at the normal rate.

Try it on your own input

Fill in an input and a few questions in the Playground and see the probabilities, then wire it in.