ML Env

Use casesVoice of customer

Summarise what your NPS promoters and detractors actually wrote

Summarises the free-text answers behind your NPS score, band by band — what promoters praise and detractors cite, backed by direct quotes.

How it works
Most businesses track the NPS number and never systematically read the free-text 'why did you give that score?' answers. This job splits the verbatims by score band, and the large model summarises each band separately: what promoters praise, what detractors cite, what passives are on the fence about, each point backed by direct quotes. It runs after each survey wave and produces a two-page brief instead of a folder of raw exports.
Data you need
NPS or CSAT survey exports with the score and the verbatim text paired per respondent — Typeform, Google Forms, Delighted and Trustpilot invitations all export this. You need enough verbatims per wave to be worth summarising; with fifteen responses, just read them yourself.
What to expect
Good at turning a wave of verbatims into something the whole team will actually read, with quotes to check against. It summarises what it sees and will not warn you that a small sample makes a theme look bigger than it is. Verbatim quality varies wildly — many respondents write three words, and the summary can end up dominated by the few who wrote paragraphs. It tells you what people said, not why the number moved.
Where people stay involved
The owner or ops lead reads the brief and checks the quotes against context before acting; the model cannot distinguish a detractor complaint about something you control from one about the courier's strike last month.

Which model, and what it costs to run

Qwen3-8B

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

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

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 voice of customer

All 15 →

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.