ModelVerdict

Austin Landscaping Inc. · INV-17836

English (US)Scan11 fields

Document en-p0034 · layout en-L06 · 11 fields · public part of the dataset · Same invoice in Czech → · Report an error in this example

Austin Landscaping Inc. · INV-17836

Verdict

Cheapest without an error: GPT-6 Luna at $0.49 per 1,000 documents.

  • GPT-6 Sol is also error-free but 19.1× more expensive.
  • Gemma 4 31B is 2.9× cheaper, but: wrong value (Supplier ID (EIN, IČO…)).
  • DeepSeek V4.1 Flash is 1× cheaper, but: wrong value (Bank account).

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

FieldCorrect valueGemma 4 31B9 / 11 fieldsDeepSeek V4.1 Flash10 / 11 fieldsGPT-6 Luna11 / 11 fieldsMistral Small 410 / 11 fieldsGemini 3.5 Flash Lite11 / 11 fieldsQwen3.8 Flash11 / 11 fieldsGemini 3.8 Flash11 / 11 fieldsMistral Medium 3.511 / 11 fieldsClaude Sonnet 511 / 11 fieldsGPT-6 Sol11 / 11 fields
SupplierAustin Landscaping Inc.Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)Austin Landscaping Inc. (correct)
Supplier ID (EIN, IČO…)33-3639325 36-3839325wrong value33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)33-3639325 (correct)
Invoice numberINV-17836INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)
Payment referenceINV-17836INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)INV-17836 (correct)
Invoice date2025-07-252025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)2025-07-25 (correct)
Due date2025-08-012025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)2025-08-01 (correct)
Net by tax rate0 %: 341.46, 7 %: 3437.720 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)0 %: 341.46, 7 %: 3437.72 (correct)
Tax by rate0 %: 0.00, 7 %: 240.647 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)7 %: 240.64 (correct)
Total4019.824019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)4019.82 (correct)
CurrencyUSDUSD (correct)USD (correct)USD (correct)USD (correct)USD (correct)USD (correct)USD (correct)USD (correct)USD (correct)USD (correct)
Bank account021000021 946505175 021000021 945803173wrong value 021000021 946585175wrong value021000021 946505175 (correct) 0211000021 946505175wrong value021000021 946505175 (correct)021000021 946505175 (correct)021000021 946505175 (correct)021000021 946505175 (correct)021000021 946505175 (correct)021000021 946505175 (correct)
Tokensinput / output584 / 2201,549 / 3582,896 / 2532,693 / 1781,406 / 2192,680 / 1,5311,406 / 1,0712,693 / 1903,119 / 2082,896 / 208
of which hidden reasoningbilled as output0222141001,3188520096
of which from cacheinput billed at a discount3074360003840000
Cost of this documentthis run$0.0002$0.0005$0.0005$0.0005$0.0010$0.0011$0.0051$0.0055$0.0083$0.0093
Per 1,000 documentsfor the same kind of document$0.17$0.48$0.49$0.51$0.97$1.07$5.07$5.46$8.32$9.32
Latencythis run11.7 s4.1 s3.3 s2.3 s1.9 s19.7 s4.2 s3.9 s3.4 s5.1 s