ML Env

Use casesVoice of customer

A Monday-morning digest of what customers complained about last week

A one-page weekly digest of support tickets — top issues, movement against last week, anything unusual — in the inbox before Monday stand-up.

How it works
During the week the small model tags each closed support ticket with topic, product and sentiment. On Sunday night the large model turns the week's tags plus a sample of raw tickets into a one-page digest: top issues, movement versus the previous week's counts, anything unusual, with example tickets linked. It lands in the inbox or Slack before the Monday stand-up.
Data you need
Support tickets exported or pulled via API from wherever they live — Zendesk, Freshdesk, Gorgias, a shared mailbox. It needs the message text and dates; agent notes and resolution codes make it noticeably better. If support happens across three personal inboxes and WhatsApp, consolidating that comes first.
What to expect
Works well on a steady ticket volume flowing through one system. On a small volume, week-over-week comparisons are mostly noise and the digest degrades into a list. It summarises what customers wrote, which is not always the actual root cause the agent discovered, and ticket threads with long quoted histories need cleaning or the model ends up summarising the boilerplate.
Where people stay involved
The support lead sanity-checks the digest against their own sense of the week — they were in the tickets, the digest is for everyone who wasn't. Week-over-week 'trends' on a modest ticket volume need a human to judge whether they are noise.

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.