Wanda Patel · 1951
PDF text layer (the model received only this text)
SELLER
Wanda Patel Invoice
2794 Riverside Drive
Columbus, OH 43216
1951
Tel. (923) 555-0131
Date 04/12/2025
BUYER Terms Net 14
Blair-Tyler Group Inc. Due date 04/26/2025
260 Elm Street
Suite 906 Pay by ACH
Atlanta, GA 30335
Bank JPMorgan Chase Bank,
N.A.
Account name Wanda Patel
Routing number 021000021
Account number 16147561536
Description Qty Unit Unit price Amount
Copy paper 8.5 x 11, case of 10 reams 1 case 56.99 56.99
Scissor lift rental 2 day 194.94 389.88
Tri-fold brochures, full color 4791 ea 0.85 4,072.35
Nitrile gloves, box of 100 15 box 14.14 212.10
No sales tax charged: services are non-taxable. Subtotal 4,731.32
Thank you for your business!
Total 4,731.32
Total due: $4,731.32
Verdict
Cheapest without an error: Mistral Small 4 at $0.15 per 1,000 documents.
- GPT-6 Sol is also error-free but 25.3× 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 | Mistral Small 411 / 11 fields | Gemma 4 31B11 / 11 fields | GPT-6 Luna11 / 11 fields | DeepSeek V4.1 Flash11 / 11 fields | Gemini 3.5 Flash Lite11 / 11 fields | Qwen3.8 Flash11 / 11 fields | Mistral Medium 3.511 / 11 fields | Gemini 3.8 Flash11 / 11 fields | Claude Sonnet 511 / 11 fields | GPT-6 Sol11 / 11 fields |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Supplier | Wanda Patel | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) | Wanda Patel (correct) |
| Supplier ID (EIN, IČO…) | — | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) |
| Invoice number | 1951 | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) | 1951 (correct) |
| Payment reference | — | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) |
| Invoice date | 2025-04-12 | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) | 2025-04-12 (correct) |
| Due date | 2025-04-26 | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) | 2025-04-26 (correct) |
| Net by tax rate | — | 0 %: 4731.32 (correct) | 0 %: 4731.32 (correct) | — (correct) | — (correct) | 0 %: 4731.32 (correct) | : 4731.32 %: 0.00 (correct) | 0 %: 4731.32 (correct) | 0 %: 4731.32 (correct) | 0 %: 4731.32 (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 | 4731.32 | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (correct) | 4731.32 (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 | 021000021 16147561536 | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) | 021000021 16147561536 (correct) |
| Tokens | input / output | 665 / 135 | 702 / 185 | 699 / 237 | 864 / 474 | 690 / 185 | 1,026 / 2,559 | 665 / 135 | 690 / 779 | 853 / 183 | 699 / 233 |
| of which hidden reasoning | billed as output | 0 | 0 | 148 | 390 | 0 | 2,387 | 0 | 651 | 16 | 144 |
| of which from cache | input billed at a discount | 246 | 309 | 0 | 438 | 0 | 512 | 0 | 0 | 0 | 0 |
| Cost of this document | this run | $0.0001 | $0.0002 | $0.0002 | $0.0004 | $0.0007 | $0.0013 | $0.0020 | $0.0034 | $0.0035 | $0.0037 |
| Per 1,000 documents | for the same kind of document | $0.15 | $0.17 | $0.19 | $0.41 | $0.67 | $1.29 | $2.01 | $3.44 | $3.54 | $3.73 |
| Latency | this run | 1.2 s | 7.3 s | 2.4 s | 6.8 s | 0.9 s | 33.1 s | 1.1 s | 3.1 s | 2.8 s | 4 s |