ModelVerdict

Nathan Gonzalez

Anglais (US)E-mail9 champs

Document en-p0032 · mise en page signature · 9 champs · partie publique du jeu de données · Même facture · Tchèque → · Signaler une erreur dans cet exemple

Texte de l'e-mail (le modèle a reçu exactement ce texte)

From: Nathan Gonzalez <nathan.gonzalez@lyons-ochoa.example>
To: support@nguyen-hanson.example
Date: Wed, Mar 5, 2025 at 12:54 PM
Subject: Where is our shipment?

Hi,

Our order still hasn't arrived, even though it was supposed to be delivered last week. Order #ORD-2025-9556.

Many thanks, we really enjoy working with your team. We need an answer by the end of the month.

Thanks,

Nathan Gonzalez
Office Manager
Lyons-Ochoa
4563 John Glens Suite 670
East Jonathanberg, MO 71976
Phone: (430) 734-7900
EIN: 86-5292820

This email and any attachments are confidential and intended solely for the addressee.

Verdict

Le moins cher sans erreur : Gemma 4 31B, à 0,12 € pour 1 000 documents.

  • Claude Sonnet 5 ne fait aucune erreur non plus, mais coûte 31,5× plus cher.
  • Mistral Small 4 est 1,4× moins cher, mais : valeur erronée (Tonalité).

Prompt soumis à chaque modèle

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.

Notre notation

Correspondance exacte par champ, après normalisation des dates, des montants et des espaces. Les diacritiques manquants comptent comme une erreur. Le coût correspond à la consommation réelle de tokens de cette exécution, convertie au taux indiqué en bas de page.

Ce que chaque modèle a extrait

ChampValeur correcteMistral Small 48 / 9 champsGemma 4 31B9 / 9 champsGPT-6 Luna9 / 9 champsQwen3.8 Flash9 / 9 champsDeepSeek V4.1 Flash9 / 9 champsGemini 3.5 Flash Lite9 / 9 champsMistral Medium 3.58 / 9 champsGemini 3.8 Flash9 / 9 champsGPT-6 Sol9 / 9 champsClaude Sonnet 59 / 9 champs
Catégoriedeliverydelivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)delivery (correct)
Prioriténormalnormal (correct)normal (correct)normal (correct)normal (correct)normal (correct)normal (correct) highvaleur erronéenormal (correct)normal (correct)normal (correct)
Tonalitépositive neutralvaleur erronéepositive (correct)positive (correct)positive (correct)positive (correct)positive (correct)positive (correct)positive (correct)positive (correct)positive (correct)
Nom du clientNathan GonzalezNathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)Nathan Gonzalez (correct)
Numéro de commandeORD-2025-9556ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)ORD-2025-9556 (correct)
Numéro de facture—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Montant—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Devise—— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)— (correct)
Échéance2025-03-312025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)2025-03-31 (correct)
Tokensentrée / sortie637 / 77673 / 78749 / 153992 / 367888 / 257661 / 98637 / 77661 / 328749 / 99922 / 211
dont raisonnement cachéfacturé comme sortie0091267200002303789
dont depuis le cacheentrée facturée à prix réduit406453076858600000
Coût de ce documentcette exécution0,0001 €0,0001 €0,0001 €0,0002 €0,0002 €0,0004 €0,0014 €0,0016 €0,0023 €0,0036 €
Pour 1 000 documentspour ce même type de document0,08 €0,12 €0,14 €0,20 €0,22 €0,41 €1,41 €1,59 €2,29 €3,64 €
Latencecette exécution0,8 s5,9 s2,8 s4,9 s3,1 s0,7 s0,9 s1,7 s4,5 s3,9 s