ModelVerdict

Michaela Strnadová · 25600363

CzechScan13 fields

Document cs-p0012 · layout cs-L06 · 13 fields · public part of the dataset · Same invoice in English (US) → · Report an error in this example

Michaela Strnadová · 25600363

Verdict

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

  • GPT-6 Sol is also error-free but 20.3× more expensive.
  • Gemma 4 31B is 2.7× cheaper, but: wrong value (Supplier ID (EIN, IČO…)).

Prompt given to every model

Z přiloženého dokladu vytěž údaje a vrať pouze JSON s těmito klíči:

supplier_name      název dodavatele (přesně jak je na dokladu, včetně diakritiky a právní formy)
supplier_id        IČO dodavatele
vat_id             DIČ dodavatele
invoice_number     číslo dokladu
payment_reference  variabilní symbol
issue_date         datum vystavení (YYYY-MM-DD)
tax_point_date     datum uskutečnění zdanitelného plnění, DUZP (YYYY-MM-DD)
due_date           datum splatnosti (YYYY-MM-DD)
net_by_rate        základ daně podle sazeb DPH, např. {"21": 18450.00, "12": 1200.00}
vat_by_rate        DPH podle sazeb, např. {"21": 3874.50}
total              celkem k úhradě (číslo, záporné u dobropisu)
currency           měna podle ISO 4217 (CZK, EUR, …)
bank_account       číslo účtu dodavatele (tvar předčíslí-číslo/kód banky) nebo IBAN

Pravidla:
- Údaje odběratele nevracej, jen dodavatele.
- Pokud údaj na dokladu není, vrať null. Nic si nedomýšlej.
- Pokud doklad neobsahuje rozpis DPH (neplátce, zálohová faktura), vrať net_by_rate a vat_by_rate jako null.
- Částky jako čísla s desetinnou tečkou, bez oddělovačů tisíců a bez měny.

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 31B12 / 13 fieldsGPT-6 Luna13 / 13 fieldsMistral Small 412 / 13 fieldsDeepSeek V4.1 Flash12 / 13 fieldsQwen3.8 Flash13 / 13 fieldsGemini 3.5 Flash Lite13 / 13 fieldsMistral Medium 3.511 / 13 fieldsGemini 3.8 Flash13 / 13 fieldsClaude Sonnet 513 / 13 fieldsGPT-6 Sol13 / 13 fields
SupplierMichaela StrnadováMichaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)Michaela Strnadová (correct)
Supplier ID (EIN, IČO…)51923777 5193777wrong value51923777 (correct)51923777 (correct)51923777 (correct)51923777 (correct)51923777 (correct) 65014537wrong value51923777 (correct)51923777 (correct)51923777 (correct)
VAT ID—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct) CZ65014537invented value— (correct)— (correct)— (correct)
Invoice number2560036325600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)
Payment reference2560036325600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)25600363 (correct)
Invoice date2025-08-132025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)2025-08-13 (correct)
Tax point—— (correct)— (correct) 2025-08-13invented value— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Due date2025-08-202025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)2025-08-20 (correct)
Net by tax rate—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Tax by rate—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Total145624.29145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)145624.29 (correct)
CurrencyCZKCZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)CZK (correct)
Bank account4909928634/01004909928634/0100 (correct)4909928634/0100 (correct)4909928634/0100 (correct) 4989928634/0100wrong value4909928634/0100 (correct)4909928634/0100 (correct)4909928634/0100 (correct)4909928634/0100 (correct)4909928634/0100 (correct)4909928634/0100 (correct)
Tokensinput / output704 / 1983,122 / 1962,592 / 1691,746 / 2762,911 / 4961,516 / 1982,592 / 1691,516 / 1,1373,396 / 2343,122 / 213
of which hidden reasoningbilled as output08801533090098945105
of which from cacheinput billed at a discount4270021864000000
Cost of this documentthis run$0.0002$0.0005$0.0005$0.0005$0.0006$0.0009$0.0052$0.0054$0.0091$0.0099
Per 1,000 documentsfor the same kind of document$0.18$0.49$0.49$0.52$0.58$0.95$5.16$5.40$9.13$9.93
Latencythis run7.2 s2.6 s2.3 s4.2 s7.3 s1.8 s2.8 s4.3 s4.7 s4.4 s