ModelVerdict

Tyler Brown

English (US)Email9 fields

Document en-p0016 · layout thread · 9 fields · public part of the dataset · Same invoice in Czech → · Report an error in this example

Email text (the model got exactly this text)

From: Tyler Brown <tyler.brown@whitaker-anderson.example>
To: support@miller-jones.example
Date: Wed, Nov 19, 2025 at 1:58 PM
Subject: Re: Can't log in

Hi,

I'm following up on the message below, nothing has happened so far. Thank you for handling this. Our operations are at a standstill right now.

Kind regards,
Tyler Brown
Whitaker, Anderson and Miles

On Thu, Nov 13, 2025 at 1:25 PM, support@miller-jones.example wrote:

> Hello, thank you for your message. We've logged your request as ticket #92733 and are looking into it.
>
> Miller-Jones

> On Thu, Nov 13, 2025 at 11:58 AM, Tyler Brown <tyler.brown@whitaker-anderson.example> wrote:

> > Hi,
> >
> > Since this morning we can't log in to your portal, the page shows a 500 error.
> >
> > Kind regards,
> > Tyler Brown
> > Whitaker, Anderson and Miles

Verdict

No model extracted this document without an error. Closest: Gemma 4 31B (1 errors).

    Prompt given to every model

    You triage the support inbox. Read the email below and return only JSON with these keys:
    
    category        one of: billing (invoices, payments, charges), delivery (where is the order, shipping,
                    address change), complaint (damaged, wrong or defective goods, warranty claims),
                    technical (login, app or system errors, integrations), cancellation (cancel an order,
                    contract or subscription), sales (quotes, prices, interest before buying)
    priority        one of: urgent, high, normal, low (rules below)
    sentiment       one of: positive, neutral, negative (the customer's tone)
    customer_name   full name of the customer who wrote the request
    order_number    the customer's order number, if mentioned
    invoice_number  invoice number, if mentioned
    amount          the amount of money the request is about (number)
    currency        ISO 4217 code of that amount (USD, EUR, CZK, …)
    deadline        the date by which the customer wants a resolution or answer (YYYY-MM-DD)
    
    Priority rules:
    - urgent: the customer cannot work right now, or the deadline is at most 1 day after the email date
    - high: the deadline is at most 7 days after the email date, or the customer threatens to cancel
      or escalate (lawyer, ending the business)
    - low: the customer says explicitly that it is not urgent
    - normal: everything else
    
    Rules:
    - A thread: judge the newest message; order numbers, invoice numbers and amounts may come from the
      quoted earlier messages. A forwarded email: the customer is the original sender, not the
      colleague who forwarded it.
    - Resolve relative deadlines ("by Friday", "by tomorrow", "by the end of the month") against the
      date of the message that states them.
    - Support ticket numbers, tax IDs and phone numbers are not order or invoice numbers.
    - If a value is not in the email, return null. Do not guess.
    - 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 31B8 / 9 fieldsGPT-6 Luna8 / 9 fieldsMistral Small 47 / 9 fieldsDeepSeek V4.1 Flash8 / 9 fieldsQwen3.8 Flash8 / 9 fieldsGemini 3.5 Flash Lite8 / 9 fieldsMistral Medium 3.58 / 9 fieldsGemini 3.8 Flash8 / 9 fieldsGPT-6 Sol8 / 9 fieldsClaude Sonnet 58 / 9 fields
    Categorytechnicaltechnical (correct)technical (correct)technical (correct)technical (correct)technical (correct)technical (correct)technical (correct)technical (correct)technical (correct)technical (correct)
    Priorityurgenturgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)urgent (correct)
    Sentimentneutral negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value negativewrong value
    Customer nameTyler BrownTyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)Tyler Brown (correct)
    Order number—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
    Invoice number—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
    Amount—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
    Currency—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
    Deadline—— (correct)— (correct) 2025-11-20invented value— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
    Tokensinput / output736 / 57812 / 95695 / 69955 / 1591,055 / 680724 / 76695 / 59724 / 361812 / 891,021 / 103
    of which hidden reasoningbilled as output043011260300285370
    of which from cacheinput billed at a discount4005867680374000
    Cost of this documentthis run$0.0001$0.0001$0.0001$0.0002$0.0004$0.0004$0.0015$0.0019$0.0025$0.0031
    Per 1,000 documentsfor the same kind of document$0.13$0.13$0.15$0.19$0.37$0.41$1.49$1.90$2.51$3.07
    Latencythis run1.2 s2.4 s0.7 s2.1 s9.2 s0.7 s0.7 s2.2 s3.4 s2.1 s