API · Last updated 2026-09-21

Jev API score endpoint

type: "score" with an ordered 2–10 level array returns a probability-weighted score, legend, probabilities, and confidence. There is no documented /v1/score.

This page is for builders who searched an extra “endpoint” name while wiring Jev / System One. jev.pro is an independent unofficial field guide — not affiliated with TypeSafe AI. We do not sell or issue keys. Confirm first-party routes on docs.typesafe.ai/api.

The public API reference documents one evaluation route — POST https://api.typesafe.ai/v1/systemone — plus GET /v1/models. This title is not a separate first-party URL in that reference. Treat the capability as something you implement around the official call.

Steps

1. Write descriptive, ordered rungs a person could apply.
2. Threshold “past level 1” in code — do not treat 1.37 as a physical quantity.
3. Split multi-factor scores; weight in code (composite scoring).
4. Gate on confidence when the distribution is flat.

Map onto the official POST

Criteria: array of 2–10 levels. score may sit between levels. legend keys are string indexes. Jaggedness: weak numerical calibration on jev-1.13. Official docs warn against raw numbers you later interpolate. Question ids are not sent to the model.

Noul is the wrong primitive if you wanted a 0–1 yes probability without rungs. Choice is the wrong primitive if the options are ordered severity. Mix Score + Choice in one questions map; do not open a second HTTP path.

Pin jev-1.13.0 once floors exist. jev-latest and jev-preview currently resolve to it — confirm. Documented errors: 401, 422, 429, 529. Official SDKs retry 429/529. Input tokens meter at the vendor list price ($0.042 / million; output free — vendor claim). Schema-safe ≠ a calibrated physical unit. Log legend with the numeric score so a later reader knows which rungs you shipped.

When not to use Score

Do not ask Jev to multiply weights or reconstruct a dollar amount between levels. Do not use Score as a checksum. If you need yes/no, use Noul. If you need an unordered label, use Choice (max 255).

Pitfalls

• One Score that secretly mixes three dimensions.
• Asking Jev to multiply weights.
• Calling /v1/score.
• Treating confidence on Score as factual correctness.

See also

• API hub — evaluation POST
• Score questions — level-array rules
• Composite scoring — weights in code
• Classify pattern — unordered labels

FAQ

Is /v1/score documented?

No.

Does Score have confidence?

Yes. Noul does not.

Max levels?

10, per official docs.

Can score be 1.37?

Yes. Threshold in code; do not invent a physical unit.

What this page does not claim

• Official TypeSafe status, logos, or support.
• That jev.pro sells, issues, or proxies API keys.
• Extra first-party HTTP routes beyond POST /v1/systemone and GET /v1/models unless the API reference lists them.
• Invented latency SLOs, search volume, or unofficial prices. List price and rate limits are vendor claims — confirm on models.
• That a schema-constrained answer is automatically factually correct.

Disclaimer

This is an independent unofficial site and is not affiliated with TypeSafe AI; official documentation is available at https://docs.typesafe.ai. Never treat jev.pro as TypeSafe official documentation. We do not sell, issue, or proxy API keys.