David Fitzpatrick
Text (the model got exactly this text)
Employee record – Mccoy-Ward Employee: David Fitzpatrick Date of birth: June 13, 1968 SSN: 937-45-9333 Home address: 05864 King Route Apt. 319, Clarkshire, DE 03820 Personal email: david.fitzpatrick@inbox.example Direct deposit account: 690380348742 Hired as staff accountant, 90-day probation period. Reports to: David Martinez. Employer: Mccoy-Ward, EIN 73-8367760
Verdict
Cheapest without an error: Gemma 4 31B at $0.11 per 1,000 documents.
- Claude Sonnet 5 is also error-free but 37.7× more expensive.
- Mistral Small 4 is 1.1× cheaper, but: missing field (Names).
Prompt given to every model
Find all personal data of natural persons in the text below (for anonymisation) and return only JSON with these keys, each a list of strings (an empty list when there is none): person_names full names of people, each person once, as written in the text email_addresses email addresses of people phone_numbers phone numbers of people, as written postal_addresses home or delivery addresses of people, as written (street, city, ZIP) birth_dates dates of birth (YYYY-MM-DD) national_ids national identification numbers (e.g. SSN, birth numbers) bank_accounts bank account numbers of people, as written Rules: - Only data of natural persons. Company names, company registration or tax numbers (EIN, IČO, VAT), generic company addresses such as info@ or support@, order and ticket numbers are not personal data: leave them out. - Dates that are not dates of birth (meetings, deliveries, start dates) are not personal data. - Do not invent or complete values; copy them from the text.
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 46 / 7 fields | Gemma 4 31B7 / 7 fields | GPT-6 Luna7 / 7 fields | Qwen3.8 Flash7 / 7 fields | DeepSeek V4.1 Flash7 / 7 fields | Gemini 3.5 Flash Lite7 / 7 fields | Mistral Medium 3.57 / 7 fields | Gemini 3.8 Flash7 / 7 fields | GPT-6 Sol7 / 7 fields | Claude Sonnet 57 / 7 fields |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Names | David Fitzpatrick; David Martinez | David Fitzpatrickmissing field | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) | David Fitzpatrick; David Martinez (correct) |
| Email addresses | david.fitzpatrick@inbox.example | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) | david.fitzpatrick@inbox.example (correct) |
| Phone numbers | — | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) | — (correct) |
| Addresses | 05864 King Route Apt. 319, Clarkshire, DE 03820 | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) | 05864 King Route Apt. 319, Clarkshire, DE 03820 (correct) |
| Dates of birth | 1968-06-13 | 1968-06-13 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) | June 13, 1968 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) | 1968-06-13 (correct) |
| National IDs | 937-45-9333 | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) | 937-45-9333 (correct) |
| Bank accounts | 690380348742 | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) | 690380348742 (correct) |
| Tokens | input / output | 384 / 118 | 404 / 122 | 479 / 165 | 702 / 386 | 612 / 393 | 392 / 176 | 384 / 123 | 392 / 481 | 479 / 152 | 1,093 / 179 |
| of which hidden reasoning | billed as output | 0 | 0 | 73 | 215 | 297 | 0 | 0 | 373 | 60 | 33 |
| of which from cache | input billed at a discount | 219 | 249 | 0 | 512 | 368 | 0 | 0 | 0 | 0 | 0 |
| Cost of this document | this run | $0.0001 | $0.0001 | $0.0001 | $0.0002 | $0.0003 | $0.0006 | $0.0015 | $0.0021 | $0.0025 | $0.0040 |
| Per 1,000 documents | for the same kind of document | $0.10 | $0.11 | $0.13 | $0.22 | $0.32 | $0.56 | $1.50 | $2.10 | $2.48 | $3.98 |
| Latency | this run | 1.1 s | 5.4 s | 1.9 s | 8.1 s | 5.8 s | 0.9 s | 1 s | 2 s | 2.9 s | 3.2 s |