ML Env

Use casesFinance & admin

A plain-language read of a contract before you sign it

A structured plain-words summary of a draft contract — commitments, liability, termination — so your lawyer's hour is better spent.

How it works
You drop in a draft supplier agreement, lease or platform terms document and the large model produces a structured plain-words summary: what you're committing to, payment and liability terms, termination rules. If you also provide a few of your existing contracts, it lists the clauses that deviate from what your agreements usually say. It's a preparation step, so your questions to the counterparty — or your lawyer's hour — are better spent.
Data you need
The draft contract as selectable digital text (a Word file or text-based PDF); scans and photos need converting to text first, as no vision model is served. Optionally a few of your existing contracts, also as digital text, so 'unusual for us' has a baseline.
What to expect
Treat the summary as a reading aid with known blind spots, not a risk assessment. The model can miss what matters most — an absence (no liability cap, no exit clause) is harder to flag than a present clause — and cross-references to annexes get skipped if the annex isn't included. It has no view of your national law, and legal nuance in Dutch, German and other EU-language contracts is a known blind spot. Since most signed contracts in an SMB exist only as scans, the 'unusual for us' comparison is often unavailable in practice.
Where people stay involved
This never replaces the lawyer for material contracts; it replaces going into the lawyer meeting having not really read the document. You still read every flagged clause in the original, and you sign nothing on the strength of the summary alone.

Which model, and what it costs to run

Qwen3-32B

This job runs in bulk rather than to a waiting person, so size is not the constraint — we take the strongest independent score on structured output and tool calls that we may serve freely and that fits on a single card.

Licence
Apache-2.0
Weights at 4-bit
18 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
$0.12 / $0.12per M in / out · Nebius AI Studio · EU

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.

Structured output and tool calls

Berkeley Function Calling Leaderboard · v4 · 13 of 18 models measured

Measured in a sandbox, on somebody else's functions. It tells you which models are capable of the shape of the job, not which one will survive contact with your API.

GLM-4.672.38%Kimi K259.06%DeepSeek-V3.254.12%Qwen3-32B48.71%Qwen3-235B-A22B47.99%Qwen3-8B42.57%Qwen3-30B-A3B41.39%Qwen3-14B41.03%
Longer is betterFree to serveConditions apply⚠ answered under 95%

Independent measurement · UC Berkeley (Gorilla project) · board updated 2026-04-12

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.

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

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.

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.