Which SQL dialects are supported?

The tool supports PostgreSQL, MySQL 8, and SQLite. Each dialect generates appropriate type names: PostgreSQL uses SERIAL and BOOLEAN, MySQL uses INT AUTO_INCREMENT and TINYINT(1), and SQLite uses INTEGER PRIMARY KEY AUTOINCREMENT and INTEGER for booleans.

How are JSON types mapped to SQL column types?

Strings map to VARCHAR(255), long strings to TEXT. Integers map to INT (or SERIAL for id columns), floats to DECIMAL(10,2). Booleans map to BOOLEAN (PostgreSQL), TINYINT(1) (MySQL), or INTEGER (SQLite). ISO date strings (YYYY-MM-DD) map to DATE, ISO timestamp strings map to TIMESTAMP. Nested objects and arrays map to JSON (or TEXT in SQLite).

How does the tool detect primary keys?

A field named "id" is automatically converted to a primary key with auto-increment: SERIAL PRIMARY KEY (PostgreSQL), INT AUTO_INCREMENT PRIMARY KEY (MySQL), or INTEGER PRIMARY KEY AUTOINCREMENT (SQLite).

Can I generate SQL from an array of JSON objects?

Yes. Paste a JSON array and the tool uses the first element to infer the table schema. This is the most common use case when you have API response data as an array.

What are timestamp columns?

When the timestamps option is enabled, the tool adds created_at and updated_at columns with database-level defaults. In MySQL, updated_at uses ON UPDATE CURRENT_TIMESTAMP for automatic updates. This is a common convention for tracking record creation and modification times.

How are nested JSON objects handled?

Nested JSON objects and arrays are stored as JSON columns (or TEXT in SQLite). SQL does not natively support nested tables from a single field — if you need to model nested objects relationally, you will need to split them into separate tables manually.

Free · Browser-only · No sign-up

JSON to SQL Table Generator

Paste any JSON object or array and instantly get a CREATE TABLE SQL statement with inferred column types. Supports PostgreSQL, MySQL 8, and SQLite. Detects integers, floats, booleans, dates, timestamps, and nested JSON automatically.

PostgreSQLMySQLSQLiteAuto type inferenceTimestamps100% offline

table name

timestamps

json input

sql output · PostgreSQL

Type mapping

string→VARCHAR(255)

long string→TEXT

integer→INT / SERIAL

float→DECIMAL(10,2)

boolean→BOOLEAN

date string→DATE

timestamp string→TIMESTAMP

object / array→JSON / TEXT

null→VARCHAR(255)

Features

Auto type inference

String, integer, float, boolean, date, timestamp — all detected automatically

Three dialects

Correct syntax for PostgreSQL, MySQL 8, and SQLite out of the box

Primary key detection

Fields named "id" become SERIAL / AUTO_INCREMENT primary keys

Timestamp columns

Optional created_at / updated_at with dialect-correct defaults

Array support

Paste an array of objects — the tool uses the first item as the schema

Browser-only

No server — your JSON and SQL stay on your device

How to use

1
Paste JSON
Copy a JSON object or array of objects into the input panel.
2
Configure
Enter a table name, select your SQL dialect, and toggle timestamp columns.
3
Generate
Click Generate SQL to produce the CREATE TABLE statement.
4
Review & run
Copy the SQL, review types and constraints, then run it in your database.

Frequently asked questions

Can I use the output directly in production migrations?

The generated SQL is a starting point. Review types (especially VARCHAR lengths), add foreign keys, indexes, and constraints appropriate to your data model before running it in production.

What if my JSON has inconsistent fields across array items?

The tool uses the first item in the array. If some objects have fields that others lack, those missing fields will not appear. Check the generated schema against your full dataset.

Does it support composite primary keys?

No. Auto-generated schemas use a single auto-increment id primary key. Composite keys must be added manually after generation.

How do I handle enum values?

String columns are generated as VARCHAR(255). If you know a field should be an enum (e.g., status with values active/inactive), manually change the column type to ENUM('active', 'inactive') in PostgreSQL/MySQL.

Related tools

JSON → TypeScript

Generate TypeScript interfaces from JSON

JSON → Zod Schema

Generate Zod validators from JSON

JSON Formatter

Format and validate JSON syntax