Answer 'how do we usually book this?' from your own accounting notes
Retrieval-backed answers from your own accounting notes and accountant emails, with a citation to where the past decision was written down.
- How it works
- The system indexes your internal bookkeeping knowledge — your accountant's emails and memos, booking conventions, VAT-treatment decisions from past years, year-end checklists. When 'which account did we use for trade-fair costs, and did we reclaim the VAT?' comes up, it finds the past answer and cites where it came from, instead of another email to the accountant or an inconsistent guess.
- Data you need
- This is the catch: written material must exist. Exported accountant correspondence, procedure notes, or memos. Many SMBs keep this knowledge in one person's head — if that's you, the honest first step is writing it down (the tool then makes it findable), and until then this case doesn't fly.
- What to expect
- It retrieves what you wrote, including advice that has since gone stale — date-stamping sources and showing them in the answer is essential, since VAT rules and thresholds change year to year. It will answer fluently even from thin sources, so 'no relevant document found' must be an allowed and visible outcome. A memory aid for what your accountant already told you, not a substitute for one.
- Where people stay involved
- For anything with tax consequences, the retrieved past advice is a starting point — your accountant confirms whether it still applies.
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
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.
Independent measurement · Vectara · board updated May 11, 2026
Structured output and tool calls
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.
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.
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.
- 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.
- 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.
- 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 finance & admin
All 15 →Pull the details off supplier invoices into your bookkeeping
Pulls supplier name, invoice number, dates, line items and VAT out of PDF invoices into records your accounting software can import.
Turn expense receipts into draft expense entries
Turns OCR'd receipt text into draft expense entries — merchant, date, amount, VAT and likely category — for the submitter and finance to confirm.
Sort expenses into your bookkeeping categories
Reads cryptic bank and card descriptions and proposes a category from your own chart of accounts, backed by your past bookings as evidence.
Match incoming bank payments to open invoices
Proposes matches between bank statement lines and open invoices when references are mistyped, names differ or one transfer covers two invoices.
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.