Live Jev API · real model output

Try Jev on a real decision.

Classify text, answer a yes/no question, or score an input with TypeSafe’s Jev model — then take the same request into your code. 10,000 input tokens now, 100,000 when you sign in.

Model

1 Text to judge

0 / 8,000

2 Questions

0 / 8

10,000 input tokens without signup.
100,000 when you sign in.

Request preview

Your draft stays in this browser tab.

Your judgment

Real Jev output

A question in. A judgment out.

Your result appears here with the response returned by Jev. Run the example to see how the question and answer options shape the judgment.

Import configuration

Model judgment, not a guarantee. Review the result before using it to trigger an action.

Reading the result

What your Jev playground result tells you.

With Choice you define the possible answers and Jev returns a judgment across them, so your code gets a named answer instead of free text. Yes/No returns the probability of “yes”. Score places the text on the scale you defined.

Read the probabilities next to the selected answer. They show how the model weighs the options for this text and question; they do not certify that the answer is right. Jev does not write replies or explain its reasoning — you decide what your application does next.

01Try the judgmentStart with the example, then edit the text, question, and answers.

02Keep the configurationExport it or copy the API example.

03Get a keySign in and create one on your account page.

04Call it from codePOST to /v1/systemone. Packs from $9.90 when you need more.

One ticket · three questions

Same support ticket.
Three useful judgments.

choice

Route the ticket

Which team should handle this support ticket?

Billing, Technical, and Account are the destinations, each with a short definition. Useful when the next step depends on a specific category.

yes/no

Check urgency

Does this message describe a problem that needs attention today?

A deadline can matter even when the message sounds polite. Ask it separately from routing so your code can treat the two decisions differently.

score

Assess frustration

How frustrated does the customer sound?

0Calm
1Frustrated but polite
2Very frustrated

Jev playground FAQ

Is this Jev playground a real API call?

Yes. Every run is sent to the hosted Jev System One API and the result you see is the live model output, including the model version and token usage. Open “Raw response” under a result to inspect it.

Can Jev write a reply or explain its answer?

No. Jev makes structured judgments; it does not generate free-form text or reasoning. Use Choice for named alternatives, Yes/No for a binary question, and Score for a defined scale.

Does a high probability mean the answer is correct?

No. A probability expresses the model’s judgment for the supplied text and question. Inspect ambiguous cases and keep your own checks before triggering an action.

Is the Jev playground free?

Guests get 10,000 input tokens (about 60 runs). Sign in for 100,000 input tokens (about 500 runs). After that, packs start at $9.90 for 15,000,000 input tokens and never expire. Successful requests charge input tokens; output is free.

Can I reuse what I tried in the browser?

Yes. Export the configuration or copy the API example, create a key on your account page, and POST the same body to https://jevmodel.org/v1/systemone. Successful requests charge input tokens; failed requests are refunded.

Are my inputs and results stored?

Your draft stays in this browser tab. We record usage and billing metadata, not your text or results. Use fictional or non-sensitive text.

Is jevmodel.org affiliated with TypeSafe?

No. Jev Model Guide is an independent site. The official TypeSafe console is at console.typesafe.ai.

New to Jev? Read what Jev is, the API request shapes, or how to use Jev.