Procure-to-pay · finance workload

extract.invoice.v1

Extract structured data (vendor, dates, currency, line items, totals) from 15 synthetic invoices with planted edge cases: multi-page, credit note, foreign currency, discount lines.

No published results for this workload yet — results appear here with the next edition.

Rate-card unit

Process 1,000 invoices — rates on the rate card are the cost per successful outcome scaled to this unit.

Pass criteria

All required fields exact-match after normalization (ISO dates, decimal amounts, case-insensitive strings); line items must match in full. Grading is deterministic; a model output either passes or fails, and infrastructure errors are excluded and rerun rather than counted against the model.

Prompt (verbatim)

Every model receives exactly this prompt with the item's fields substituted for the {placeholders} — no model-specific tuning.

Extract structured data from the invoice text below.

Respond with ONLY a JSON object, no other text, using exactly this schema:
{{"vendor": str, "invoice_number": str, "invoice_date": "YYYY-MM-DD",
  "due_date": "YYYY-MM-DD", "currency": "ISO 4217 code",
  "line_items": [{{"description": str, "quantity": int,
                   "unit_price": "decimal string", "amount": "decimal string"}}],
  "subtotal": "decimal string", "tax": "decimal string", "total": "decimal string"}}

For credit notes, amounts are negative.

Invoice text:
{text}

Dataset

Datasets are generated synthetically from seeded generators; the seed is frozen per edition and held out until the run window closes. Generator source hash (SHA-256): 44c28ff33ad3dea8c26b34d4b9881747685a09676f95684f0c3d9bda73f4a195. See the methodology §7 for the verifiability protocol and any edition page for the raw graded logs.