Jev

Name: Jev
Brand: TypeSafe
Availability: InStock

Decision modelFree

typesafe/jev

one calibrated yes/no, choice or score a script can branch on; answers with probabilities, never text.

Not a chat model: it runs Headless and Provider API only. It is not in /model and cannot drive an interactive session.

Input

$0.00 /M

Output

$0.00 /M

Plan

GOAT and above

Answers with

probabilities

Runs on

Headless · Provider API

never in /model

Question types

noul · choice · score

as the request spells them

Context window

32K tokens

state plus the longest question

Modalities

→probabilities

deal spotlight

Every request on Jev is billed $0.00 — input, output and cache reads alike.

FREE

ends September 24, 2026

Full terms →

Input $/M

$0.00

was $0.042

Output $/M

$0.00

Cache read $/M

$0.00

run it

The request is the prompt: a state (text or JSON) and named questions, each with a type and instructions. Both surfaces take the same shape and return the same answers.

headless

cmd -p '{"state": "Payments failed for three days.", "questions": {"urgent": {"type": "noul", "instructions": "Does this need urgent attention?"}}}' -m typesafe/jev

Pipe the JSON on stdin instead of quoting it, and add --output-format json for the one-line result envelope.

provider api

curl https://api.commandcode.ai/provider/v1/systemone \
  -H "Authorization: Bearer <CMD_API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"model": "typesafe/jev", "state": "Payments failed for three days.",
       "questions": {"urgent": {"type": "noul", "instructions": "Does this need urgent attention?"}}}'

response

{
  "model": "typesafe/jev",
  "answers": {"urgent": {"type": "noul", "noul": 0.95}},
  "usage": {"input_tokens": 41, "output_tokens": 12}
}

noul

yes / no, as a probability that the answer is yes

choice

one of your named options, with a probability for each

score

a position on your ordered scale, with a probability for each step

Full request and response reference: Headless docs · Provider API docs

pricing

Billed per input token, at cost, on every plan that can call it. Output tokens are free. Requests count against the same credits as any other model.

Input

$0.00 / M

list $0.04 / M

Output

$0.00 / M

answers are never billed

Cache

—

no prompt cache: every request is one fresh state

frequently asked

How much does Jev cost?

Free, ends September 24, 2026: every request bills $0.00. After that, $0.04/M input tokens at cost; everything else is free.

Which plan do I need?

Available on GOAT and above, the plans with Provider API access.

How do I run Jev?

Headlessly: pass the request as JSON to cmd -p with -m typesafe/jev. From your own code: POST it to /provider/v1/systemone on the Provider API with your Command Code key. It is not in /model and cannot drive an interactive session.

Why is Jev not in /model?

Jev is a decision model, not a chat model. It answers typed questions about a state with probabilities and never produces text, so it has no place in the agent loop or a session. It is the model a script calls when it needs one calibrated judgment to branch on.

Headless docs→

how to call Jev from the CLI

Provider API docs→

how to call Jev from your own code

All supported models→

sortable by score, speed and price