Derek Richardson · INV-16398
PDF text layer (the model received only this text)
I nvoIce
number INV-16398
Supplier: Customer:
Derek Richardson Keystone Supply LLC
8074 Park Avenue 5037 Maple Drive
Suite 498 Suite 313
Chicago, IL 60688 Seattle, WA 98158
Tel. (540) 555-0144
Invoice date May 18, 2025
Terms Net 14
Due date June 1, 2025
P.O. number PO-39163
Description Qty Unit Unit price Amount
Laptop docking station 1 ea 211.99 211.99
Drywall sheet 1/2 in. 4 x 8 ft 66 ea 16.99 1121.34
Subtotal 1333.33
Total 1333.33
Total payable: USD 1333.33
Pay by ACH
Remit payment Bank PNC Bank, N.A.
Account name Derek Richardson
Routing number 043000096
Account number 2348829041
Payment reference CUST-14200
No sales tax charged: services are non-taxable.
Thank you for your business!
Authorized signature: ____________________
Verdict
Cheapest without an error: GPT-6 Luna at $0.16 per 1,000 documents.
- GPT-6 Sol is also error-free but 26.4× more expensive.
Prompt given to every model
Extract the following data from the attached document and return only JSON with these keys:
supplier_name supplier's name (exactly as printed, including the legal form)
supplier_id supplier's tax ID (EIN), if printed
invoice_number document number
payment_reference payment reference to quote with the payment
issue_date invoice date (YYYY-MM-DD)
due_date due date (YYYY-MM-DD)
net_by_rate amount per sales tax rate, e.g. {"8.25": 1845.00}; non-taxable amounts under "0"
vat_by_rate sales tax per rate, e.g. {"8.25": 152.21}
total total due (number, negative for credit notes)
currency ISO 4217 currency code (USD, GBP, EUR, …)
bank_account supplier's bank details as "routing-number account-number", or IBAN
Rules:
- Return the supplier's data only, never the customer's.
- If a value is not on the document, return null. Do not guess.
- If the document has no tax breakdown, return net_by_rate and vat_by_rate as null.
- Amounts as numbers with a decimal point, no thousands separators and no currency symbol.
How we score
Exact match per field after normalising dates, amounts and whitespace. Missing diacritics count as an error. Cost is the real token spend of this run, converted at the rate shown in the footer.
What each model extracted
| Field | Correct value | GPT-6 Luna11 / 11 fields | Gemma 4 31B11 / 11 fields | Mistral Small 411 / 11 fields | DeepSeek V4.1 Flash11 / 11 fields | Gemini 3.5 Flash Lite11 / 11 fields | Mistral Medium 3.511 / 11 fields | Qwen3.8 Flash10 / 11 fields | Gemini 3.8 Flash11 / 11 fields | Claude Sonnet 511 / 11 fields | GPT-6 Sol11 / 11 fields |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Supplier | Derek Richardson | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) | Derek Richardson (correct) |
| Supplier ID (EIN, IČO…) | — | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) |
| Invoice number | INV-16398 | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) | INV-16398 (correct) |
| Payment reference | CUST-14200 | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) | CUST-14200 (correct) |
| Invoice date | 2025-05-18 | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) | 2025-05-18 (correct) |
| Due date | 2025-06-01 | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) | 2025-06-01 (correct) |
| Net by tax rate | — | — (correct) | 0 %: 1333.33 (correct) | 0 %: 1333.33 (correct) | — (correct) | 0 %: 1333.33 (correct) | 0 %: 1333.33 (correct) | : 1333.33 %: 1333.33invented value | 0 %: 1333.33 (correct) | 0 %: 1333.33 (correct) | — (correct) |
| Tax by rate | — | — (correct) | 0 %: 0.00 (correct) | — (correct) | — (correct) | 0 %: 0.00 (correct) | — (correct) | — (correct) | 0 %: 0.00 (correct) | — (correct) | — (correct) |
| Total | 1333.33 | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) | 1333.33 (correct) |
| Currency | USD | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) | USD (correct) |
| Bank account | 043000096 2348829041 | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) | 043000096 2348829041 (correct) |
| Tokens | input / output | 676 / 175 | 681 / 160 | 634 / 146 | 847 / 384 | 669 / 194 | 634 / 146 | 993 / 1,235 | 669 / 570 | 868 / 201 | 676 / 274 |
| of which hidden reasoning | billed as output | 80 | 0 | 0 | 293 | 0 | 0 | 1,050 | 412 | 25 | 179 |
| of which from cache | input billed at a discount | 0 | 309 | 0 | 436 | 0 | 0 | 768 | 0 | 0 | 0 |
| Cost of this document | this run | $0.0002 | $0.0002 | $0.0002 | $0.0003 | $0.0007 | $0.0020 | $0.0026 | $0.0026 | $0.0037 | $0.0041 |
| Per 1,000 documents | for the same kind of document | $0.16 | $0.16 | $0.18 | $0.35 | $0.69 | $2.05 | $2.59 | $2.64 | $3.75 | $4.09 |
| Latency | this run | 1.9 s | 7.3 s | 1.1 s | 4.3 s | 1 s | 1.1 s | 20.4 s | 2.5 s | 2.3 s | 6 s |