ML Env

Use casesHR & internal

Let staff ask the handbook a question instead of asking HR

A private chat box that answers staff policy questions from your own handbook, with a link to the exact section. Runs entirely in Amsterdam.

How it works
Your handbook and policy documents are indexed with multilingual embeddings. When an employee types a question — "how many days before I have to submit a holiday request?" — the system retrieves the most relevant passages and the small model writes an answer strictly from those passages, with a link to the exact section it came from. Everything runs on our hardware in the Amsterdam data centre; nothing goes to a third party.
Data you need
A written staff handbook and policy documents (PDF, Word or wiki pages) that are actually up to date. If your policies live in people's heads rather than in documents, there is nothing to retrieve — writing them down is the real first step.
What to expect
Reliable for the questions HR answers over and over, because each answer is pinned to a retrieved passage rather than the model's imagination. Answers are only as current as the documents — a stale handbook produces confidently stale answers — and ambiguous policies get ambiguous answers. It cannot see personal data such as remaining holiday days or salary unless you separately connect your HR system, which most small companies should not rush into.
Where people stay involved
HR stays the authority for anything binding: leave balances, contract terms, disciplinary matters. The bot is set up to say "check with HR" for personal cases, because it knows the general policy, not the individual's contract or accrued leave.

Which model, and what it costs to run

Qwen3-8B

This job answers while someone waits, so latency comes first: we hold it to a single small card and, within that, take the best independent score on sticking to the document.

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.

Sticking to the document

Vectara Hallucination Leaderboard · HHEM · 17 of 18 models measured

Measured on public documents, by a model acting as judge. Read it beside the answer rate: the lowest hallucination rates on this board belong to models that simply decline more often.

Phi-43.7% · answers 80.7% ⚠Llama 3.3 70B4.1% · answers 99.5%Gemma 3 12B4.4% · answers 97.4%Qwen3-8B4.8% · answers 99.9%Mistral Small 3.25.1% · answers 97.9%Granite 4.0 Small5.2% · answers 100%DeepSeek-V3.25.3% · answers 96.6%Qwen3-14B5.4% · answers 99.9%
Shorter is betterFree to serveConditions apply⚠ answered under 95%

Independent measurement · Vectara · board updated May 11, 2026

Reading a long document whole

No benchmark measures this

No current benchmark ranks today's open models here. HELMET showed that the popular test — finding a planted sentence — predicts nothing, and its own table has not been rerun on 2026 models.

So we do not show a chart here. We measure it on your own content, in the first week, and you see the result before anything ships.

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 hr & internal

All 14 →

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.