API · Last updated 2026-09-21

Jev API extract endpoint

Date-extraction and pre-parsed value cookbooks keep arithmetic and regex in code. Jev picks among spans or names parts; code normalizes. There is no documented /v1/extract that returns emails as generated strings.

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. Extract candidates with regex, parsers, or a generative model.
2. Ask Jev Choice among candidates or Noul “is this the payout date?”
3. Normalize in code. jev-1.13 is weak at numeric precision (vendor jaggedness note).
4. Do not chain Choices to spell a string.

Map onto the official POST

Same POST. Candidates belong in state or structured instructions. Choice is unordered, max 255. Noul has no confidence field. Score must not reconstruct a dollar amount between levels. Question ids are not sent to the model.

Do not invent an extraction JSON schema for Jev to fill like an LLM. Open string fields are structured-output LLM territory. Pin jev-1.13.0; aliases currently resolve there — confirm. Documented errors: 401, 422, 429, 529. Official SDKs retry 429/529.

Vendor-claimed context (64k / 32k) argues for candidate lists, not the raw PDF binary. Text only — extract text before state. Schema-safe “this span is the date” ≠ the date is correct in the world. Official cookbooks: pre-parsed value extraction; date extraction. Read them at the source.

When not to use this pattern

If you need a free-form email or address string with no candidate set, you are outside System One. If you need arithmetic totals, keep them in code.

Pitfalls

• Using Score to reconstruct a dollar amount between levels.
• Inventing an extraction JSON schema for Jev to fill like an LLM.
• Calling /v1/extract.
• Chaining Choices to spell a string.

See also

• API hub — judgment POST
• Jaggedness — numeric weakness
• Cite pattern — support vs generate
• Choice questions — pick among spans

FAQ

Is /v1/extract documented?

No.

Official cookbooks?

TypeSafe: pre-parsed value extraction; date extraction. Read them at the source.

Need open string fields?

That is structured-output LLM territory — comparison.

Can Jev return a generated email?

Not as a System One primitive. Pick among candidates you already have.

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.