ML Env

Use casesSelling across languages

Sort every incoming message by language and topic, automatically

Detects each message's language and topic and routes it to the right queue — French returns to the French speaker, urgent complaints to the front.

How it works
The small model reads each incoming email, chat or contact-form message, detects the language, tags the topic — where-is-my-order, return request, product question, complaint, supplier — and routes it to the right queue or person. Your categories go straight into the prompt as a written list with one-line definitions and a few examples; no training run is needed. It triages in moments per message; it does not answer anything.
Data you need
Your message stream and a written list of your queues and categories, with one-line definitions and a few example messages each. A labelled sample of a hundred past messages lets you measure accuracy before trusting it.
What to expect
Mixed-topic messages — a return request plus a new-order question — get one tag, so half the message risks being missed. Misrouting is cheap when a human corrects it in seconds but expensive if a legal complaint sits in the wrong queue, which is why complaint-flavoured messages should over-trigger the urgent route rather than under-trigger, and why the model needs an 'unsure' bucket rather than being forced to guess.
Where people stay involved
Humans handle every message; the model only chooses the queue. Give it an 'unsure' bucket to route to, and have someone skim its routing decisions in the first weeks.

Which model, and what it costs to run

Qwen3-8B

No public benchmark ranks this particular job, so this is not a leaderboard pick — it is where we would start: small, permissively licensed, and strong wherever it has been measured. We prove it on your own content before anything ships.

Licence
Apache-2.0
Weights at 4-bit
5 GB
Context
32K tokens
Publisher
Alibaba (Qwen Team)

What the hardware costs

One 48 GB card holds it

Rent in the EU
$1.60/hrScaleway, Paris (PAR2)
Buy the card
$7,569new, one-off
Or rent it by the token
$0.04 / $0.04per M in / out · DeepInfra

Hardware only, third-party prices from 2026-07. The figure excludes the KV cache, which grows with context length and how many people use it at once — sized properly in a conversation, not guessed here. Renting by the token is cheaper up front; why our customers still self-host is below.

European languages

openGPT-X European LLM Leaderboard · FLORES-200 translation, BLEU · 10 models

Previous-generation models — the board has not been rerun on the 2026 roster, so read the order of magnitude, not a ranking of today's models. And BLEU scores adequacy, not tone: it cannot tell you whether the register fits your brand. That is judged on your own copy.

EuroLLM-9B25.6 · EULlama 3.1 70B24.3Mixtral 8x7B17.3Mistral-Nemo 12B17.1Teuken-7B sigma17 · EULlama 3.1 8B17Llama 3 8B15.9Teuken-7B v0.615 · EUCommand-R 35B13.8Mistral-7B v0.313.3
Higher BLEU is closer to a human translationEU-sovereign model

Independent measurement · openGPT-X / Fraunhofer IAIS · captured 2026-07-10

The API is cheaper per token. Here is why our customers don't use it.

We will not pretend otherwise: renting a model by the token from a serverless API costs less per million tokens than a card we run for you. We show that price on every use-case page. What it does not include is the part a shop with a customer database actually pays for.

  1. 01

    Your data never leaves hardware you can point at

    A serverless "we don't retain your data" is a clause in a contract. Running the model on a card in Amsterdam is a fact of architecture: your catalogue, tickets and customer records are never sent to a third party at all. For a GDPR audit, that is the difference between a promise and a floor plan.

  2. 02

    The price cannot move without your say-so

    A serverless rate card is somebody else's lever. The provider can raise the price, retire the model, or change the terms, and your cost moves with it. The same model on the same card costs the same next year — you own the number.

  3. 03

    The model cannot be taken away

    Hosted APIs deprecate models on their own schedule; the one you built on can be gone in a quarter. An open-weight model on your own hardware runs for as long as you keep the lights on. No vendor can end-of-life it out from under you.

And the price gap closes with volume: past a card you keep busy — very roughly four billion tokens a month — owning is cheaper outright, even before the three reasons above.

Getting a case like this one from a conversation to production takes about two months, and you can stop at the end of any phase.

Also in selling across languages

All 16 →

Next

Tell us what your team does by hand

Describe the process that takes the most time. We will say plainly whether a model is the right tool for it.