MN-Violet-Lotus-12B-GGUF
mradermacher · Kit
Static GGUF files of FallenMerick's MN-Violet-Lotus-12B in sizes from 4.9 GB to 13.1 GB. For people who run a 12B model on their own computer.
New
Overview
- Run it
- Read the sheet
- Look up the official register
- Keep the columns worth writing back
- Write it back to the sheet
An n8n workflow that takes the companies listed in a Google Sheet, finds each in seven European business registers and fills in its registration number, status and address.
Fill a Google Sheet of companies from European registers is an n8n workflow in the Business category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.
At a glance
- Runs in
- n8n Cloud: bring in the file, connect your accounts and run it
- Your own n8n server: the same file
- A local install: it runs, because Apify and Google are reached over the internet
- A schedule: swap the manual start for a Schedule Trigger if you want it to run on its own
Platform
How it works
Trigger
A manual run
It starts when you run the workflow by hand. Set first:
- An Apify credential and the Apify node
- A Google Sheets credential, with the sheet and tab chosen in both sheet nodes
- A company column in the sheet
Actions
- The rows of the sheet are read
- An Apify Actor looks each company up in seven national registers
- Twelve fields are kept: the registration number, status, address, a link and more
- The matching row is updated in the sheet, or a new one is added
Result
A sheet whose companies carry official register data
Integrations
What you need before you start.
Google Sheets
Company list
Is read at the start and written to at the end.
NeedsGoogle Sheets credential and the sheet picked in two nodes
Apify
Register lookup
Runs the company registry Actor once for each row and returns its result.
NeedsApify API token
Workflow check
Checked by Mindket on 13 Nov 2026
- Steps
- 5 steps, starting by hand
- AI
- No step calls an AI model
- Sign-ins
- Signs in to: Google Sheets and Apify
- API keys
- No API key is left in the file
- Community nodes
- Needs the community node package @apify/n8n-nodes-apify
- Step types
- Mindket’s checker reads all 4 of its step types
About
What it is
This n8n workflow fills a Google Sheet of companies with official register data. It reads a column of company names or registration numbers and asks an Apify Actor to look each one up in the business registers of seven European countries. The registration number, status, address and more go back into the same row. The fields come back with identical names for each country, so a single set of columns serves all seven. Five nodes run it. One sticky note sits beside them.
Who it suits
Teams that keep lists of suppliers, customers or leads in a sheet and want the official facts next to each name, such as sales researchers, buyers and compliance clerks. It suits a list of a few dozen to a few hundred companies in Czechia, Estonia, Finland, France, Norway, Poland or Slovakia. Skip it for companies elsewhere, because those seven are the only registers it reads. Skip it too if you need directors or owners. The Actor returns company data only.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Sales | Lead researchers | Add the registration number and status to a prospect list | A researcher sees at once which prospects the register still shows as active. |
| Procurement | Buyers | Check a supplier list against the official record | A buyer opens each supplier’s register page from the sheet. |
| Compliance | Onboarding clerks | Confirm that a business customer exists before onboarding | A clerk compares the registered address with the contract. |
| Finance | Credit controllers | Add VAT numbers where a register publishes them | A controller fills the VAT column for a few customers in Norway and France. |
You need an n8n, an Apify account and a Google account that can open your sheet.
Where to use it
Any n8n that can reach Apify and Google will run it. The workflow starts by hand from the editor. Each lookup waits for the Actor, so a long list takes a while.
- n8n Cloud: bring in the file, connect your accounts and run it
- Your own n8n server: the same file
- A local install: it runs, because Apify and Google are reached over the internet
- A schedule: swap the manual start for a Schedule Trigger if you want it to run on its own
Setup
Download saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Import the file, add two credentials, pick your sheet in both sheet nodes and run it on a few rows.
| Need | Detail |
|---|---|
| n8n | On n8n Cloud or on a server of your own |
| Apify | An account, an API token and the Apify node in your n8n |
| Google Sheets | A Google account that can open the sheet, with an OAuth2 credential |
| Sheet | A company column in row 1, then one name or number per row |
- In n8n, create a workflow, open its menu, choose Import from File and pick
workflow.json. - Add an Apify credential with your API token to
Look up the official register. If your n8n does not list the Apify node, install the package@apify/n8n-nodes-apifyfirst. - Add a Google Sheets OAuth2 credential to
Read the sheetand toWrite it back to the sheet. Pick your sheet and its tab in both nodes. They must be the same tab. - To search one register only, put a country code before the text in your sheet, such as
no:Equinor. The codes arecz,ee,fi,fr,no,plandsk. - Try three rows first. Select
Run itand run the workflow.
How to use it
Run the workflow when the list is ready. It reads every row, looks each one up and writes the answers back. Open the sheet afterwards to see the new columns. A row that no register returns keeps its empty cells.
| To change | Where |
|---|---|
| Countries, result count, dissolved companies | The body in Look up the official register |
| Which fields come back | The list in Keep the columns worth writing back |
| A spending cap | Maximum Cost per Run (USD) in Look up the official register |
| The key column | matchingColumns in Write it back to the sheet |
Example
A buyer keeps forty suppliers in a sheet. The company column holds names such as Equinor, and a few Czech registration numbers with the prefix cz:. She runs the workflow.
Each lookup takes a few seconds. The sheet gains columns for the registered name, status, address, city and a link to the official record. A supplier that no register returns keeps empty cells, which tells her which ones to check by hand. She opens the link on one row and finds the same address as on her contract.
For newcomers and experts
If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. A business register is a government list of the companies registered in a country. An Apify Actor is a small program on Apify’s servers that n8n can start. Here it asks seven registers the same question. Try it on three rows.
For experts. The write-back finds each row by the text you typed, not by the name the register returns.
| Node | What it does | What to know |
|---|---|---|
Look up the official register | Runs the Actor once per row and waits for it | The body is built from company with JSON.stringify. The node stops with an error unless a run ends as SUCCEEDED, and a row with no result passes nothing on |
Keep the columns worth writing back | Builds twelve fields from each result | company is the Actor’s query field, so the write-back finds your row. The register’s own name sits in registeredName |
Write it back to the sheet | Updates the row that matches company, or adds one | It auto-maps every field and makes a column for each header that is missing |
Run it | Starts the run by hand | Replace it with a Schedule Trigger to run on its own |
Good to know
Lookups run one at a time, and the Actor is small and new.
- Speed: the Actor runs once per row, one after another, each waiting up to 600 seconds. Start with a few rows
- Cost: the Actor’s price list shows USD 0.00001 for each run start and each company record. An earlier list showed USD 0.01 for each. A run that finds nothing is free. Read its page before a big run
- Actor: its page showed one user and 19 runs in total on 12 November 2026, so check it before you rely on it
- Matches: with a cap of one result, a name that fits several registers returns one match from any of them. A prefix such as
no:keeps the search to one register. Poland can be searched by number only - Gaps:
onlyActivedrops companies the register marks as dissolved.websitecomes only from Norway and Finland, andvatNumberis often empty - Repairs: the original copy wrote the register’s name into
company, which could add a new row when the spelling differed. Mindket’s copy writes your own text there and keeps the register’s name inregisteredName. The saved labels of the original sheet were removed - Not run: nothing was run, so no Actor and no sheet was tried. The notes come from Apify’s public pages and the node’s package
- Your data: the text in your
companycolumn goes to Apify, and the Actor reads the official registers - Credentials: the JSON holds none
README and files
Fill a Google Sheet of companies from European registers
An n8n workflow for a Google Sheet of companies. It reads a column of names or registration numbers and looks each one
up in the official business registers of seven European countries. The registration number, status, address and more go
back into the same row.
What is inside
workflow.json: the workflow as n8n exports it, 5 nodes and 1 sticky note. It holds no credential.README.md: these notes.
What we changed from the original copy
Keep the columns worth writing backsetcompanyto the name the register returned. The write-back finds the row to
update bycompany. A name that the register spells differently would add a new row.companynow carries the Actor'squeryfield, which is the text from your sheet, and the register's
name goes into a new column,registeredName.- Both Google Sheets nodes carried the saved labels of the original author's sheet and tab. They are blank now, so you
pick your own in each node.
Before you start
- An n8n instance, on n8n Cloud or your own server.
- An Apify account and an API token. The Apify node must be in your n8n. If it is not listed, install the package
@apify/n8n-nodes-apifyfrom the community nodes settings. - A Google account that can open your sheet, and a Google Sheets OAuth2 credential in n8n.
- A sheet whose first row has a column named
company, with one company name or registration number per row below it.
Set it up
- In n8n, create a workflow, open its menu, choose Import from File and pick
workflow.json. - Add an Apify credential with your API token to
Look up the official register. - Add a Google Sheets OAuth2 credential to
Read the sheetand toWrite it back to the sheet. Pick your sheet and
its tab in both nodes. They must be the same tab. - To search one register only, put a country code in front of the text in your sheet, such as
no:Equinororcz:27074358. The codes arecz,ee,fi,fr,no,plandsk. - Try it on three rows first. Select
Run itand run the workflow.
How a run goes
Run itstarts the workflow by hand.Read the sheetreads every row of the tab. Each row becomes one item, and thecompanycolumn holds the text to
look up.Look up the official registerruns the Apify Actorgubidonius/company-registry-lookuponce for each row. The input
is that row's text as the only query, all seven countries, at most one result, and active companies only. The node
waits up to 600 seconds for each run and returns the Actor's rows.Keep the columns worth writing backbuilds twelve fields:company,registeredName,registrationNumber,vatNumber,country,registry,status,incorporatedAt,address,city,websiteandregistryUrl.Write it back to the sheetupdates the row whosecompanymatches. A field with no matching header gets a new
column on the right.
Change the look
- Which countries are searched, how many results a query may return and whether dissolved companies count: the body in
Look up the official register. - Which fields are kept: the list in
Keep the columns worth writing back. - A spending limit: Maximum Cost per Run (USD) in
Look up the official register. It is empty here. - The column that finds the row:
matchingColumnsinWrite it back to the sheet.
Good to know
- The facts below come from Apify's public pages for the Actor and from the node package, read on 12 November 2026.
Nothing was run: no Actor run, no sheet. - The Actor runs once per row, one after another. A long sheet takes a long time, so start with a few rows.
- Cost: the Actor's price list shows USD 0.00001 for each run start and for each company record. An earlier list, from
26 August 2026, showed USD 0.01 for each. A run that finds nothing is free. Read the Actor's page before a big run. - The Actor is young and small. Its page shows one user and 19 runs in total on 12 November 2026, so check it before you
rely on it. - With
maxResultsPerQueryat 1, a name that fits several registers returns one match from any of them. A prefix such
asno:keeps the search to one register. - Poland's register can be searched by number only. A name searched there returns nothing.
onlyActiveis on, so a company that the register marks as dissolved gives no row, and its sheet row stays as it was.
The same is true for a text that no register knows.- Registers fill different fields.
websitecomes only from Norway and Finland, andvatNumberis often empty. An
empty field is written as an empty cell. - A blank
companycell is sent as a blank query. Remove empty rows first. - The Actor returns company data only. It does not return directors or shareholders.
- Your data: the text in your
companycolumn goes to Apify, and the Actor reads the official registers. - Credentials: the JSON holds none.
NOTICE.md 683 B README.md 5 KB workflow.json 7 KB
What’s new
Reviews
More from Mindket
See all from MindketT5 Small
google-t5 · Kit
Turns one text into another with a small 60M-parameter T5 model that treats translation, summarization, question answering and classification as text-to-text j…
New
VGGT-1B
facebook · Kit
Infers a scene's 3D structure from one, a few or hundreds of images in a single forward pass, including camera parameters and depth maps. It suits computer vis…
New
Similar in Business
See all in Businesspplx-decider-v1.1-27b
perplexity-ai · Kit
Picks the best option for a decision from text and images, such as which team should handle a request, with a probability for each choice, for developers who b…
New
NuExtract3
numind · Kit
Reads documents as text or images and fills a JSON template with the fields you ask for, or converts the pages to Markdown. A 4B model for teams that process s…
New
Drex DLM
nace-ai · Kit
Scores yes or no, multiple choice and rating questions about a document in one pass, with a probability for every option. An 8B decision model from Nace.AI tha…
New