Synthetic Data Generator
Generate synthetic datasets from a JSON Schema using Faker IT integrated for Italian names, addresses, tax codes and VAT numbers. Export to JSON, CSV or SQL - entirely on the client side with no Large Language Model (LLM) involved and without sending any data to the server.
Configuration
Supported: string, number, integer, boolean, object, array. Format: name | surname | email | phone | address | city | company | lorem | date-past | date-future | uuid | codiceFiscale | partitaIva. Nested $ref and oneOf not supported.
How it works
JSON Schema
Choose a preset or write your own JSON schema using types, formats, min/max values and enums for arrays and nested objects.
Configure and generate
Choose location (IT/EN/DE/FR/ES), number of records (1-1000) and data format for export. Click Generate Dataset.
Export
Download the data set as a JSON file, CSV or an SQL insert statement. Everything happens within your web browser; no data is sent to our servers.
Why use this tool?
Names, surnames, city, fiscal code, VAT number with check digit algorithm, ZIP Code, province.
Use the same scheme as your OpenAPI or Pydantic model. No proprietary format to learn.
No accounts, no fees, and no API keys are required for the most part; mostly AI and Gretel AI require a connection to the Cloud with credentials.
Come utilizzare Generatore Dati Sintetici
Choose or write a template
Select a preset (User, Product...) or write your JSON Schema with type: object and properties you want to generate.
Configure local settings, quantity and format
Choose location (Faker values), record count (1-1000), and export format (JSON, CSV or SQL).
Generate dataset
Click "Generate Dataset": The internal Faker engine creates records in the browser, without sending anything to the server.
Copy or download
Use Copy for pasting into notes or Download to save generated JSON, CSV or SQL files.
Suggerimenti
- Use local IT for quickly populating test forms or databases with valid Italian fiscal codes and VAT invoices in the correct format but not actual.
- Reuse the same schema from your OpenAPI or Pydantic model: it's not worth learning a proprietary format.
- To populate SQL tables, select the SQL format and set table name before generating, so you get ready-to-execute INSERT statements.
Domande frequenti
Are generated data realistic?
Data is plausible but completely synthetic, generated by an internal Faker generator with Italian dictionaries (names, surnames, cities, tax codes, VAT numbers with a control algorithm). It does not come from real people.
What JSON Schema formats are supported?
Types string, number, integer, boolean, object and array are supported, with formats name, surname, email, phone, address, city, company, lorem, date-past, date-future, uuid, codiceFiscale, partitaIva, cap and province. Schemas with $ref or oneOf nested are not supported.
What is the maximum number of records that can be generated?
Limit to 1000 records per generation to keep the tool fast and completely client-side without blocking the browser.
Are data being sent to a server?
No. All generations occur in the browser via JavaScript: no schema or generated data is sent to the server, and it's not involved with any AI or LLM models.
In what format can I export data?
You can export to JSON, CSV or SQL (customizable table name instructions with INSERT, download the file or copy the text into notes).