SQL CREATE TABLE Generator

Build a ready-to-run CREATE TABLE statement by describing your columns in plain text — no SQL boilerplate to remember. Pick a dialect (MySQL, PostgreSQL, SQLite, or SQL Server) and the generator handles the identifier quoting, type mapping, primary/foreign keys, auto-increment syntax, and indexes for that engine. Everything runs in your browser; nothing you type is sent anywhere.

How to use it

1. Choose your SQL Dialect. This drives every dialect-specific decision below.
2. Enter a Table Name.
3. Describe your Columns, one per line (see the syntax below).
4. Optionally add Indexes, one per line.
5. Toggle Add IF NOT EXISTS and Pretty-print output to taste.

The generated SQL updates as you type, and you can copy or download it as a .sql file.

Column syntax

Each line defines one column:

name type[(length)] [flags...]

• name — the column name (first token).
• type — one of: integer, bigint, smallint, varchar, text, boolean, date, datetime, timestamp, json, decimal, float, double, uuid, blob. Add a length/precision in parentheses, e.g. varchar(255) or decimal(10,2).
• flags (any order, space-separated):
  • pk — mark as PRIMARY KEY
  • ai — auto-increment (rendered per dialect: AUTO_INCREMENT, SERIAL/BIGSERIAL, AUTOINCREMENT, or IDENTITY(1,1))
  • notnull — NOT NULL
  • unique — UNIQUE
  • default=VALUE — a DEFAULT clause; the value is inserted verbatim (default=0, default=CURRENT_TIMESTAMP, default='active')
  • fk=table(col) — a FOREIGN KEY referencing another table's column

Example:

id bigint pk ai notnull
name varchar(255) notnull
email varchar(255) notnull unique
author_id bigint fk=users(id)
created_at timestamp notnull default=CURRENT_TIMESTAMP

Lines that are blank or start with -- are ignored, so you can leave comments.

Index syntax

Each line defines one index:

[unique] [index_name:] col1, col2

• Prefix with unique for a unique index.
• Add name: before the columns to name the index; otherwise a name is derived from the table and columns.
• List one or more columns separated by commas.

For MySQL, indexes are emitted inline inside the CREATE TABLE; for the other dialects they become separate CREATE INDEX statements.

Dialect differences

The generator adapts to each engine automatically, including:

• Identifier quoting — backticks for MySQL, [brackets] for SQL Server, and "double quotes" elsewhere.
• Type mapping — e.g. varchar becomes TEXT in SQLite and NVARCHAR in SQL Server; json becomes JSONB in PostgreSQL; boolean becomes BIT in SQL Server.
• Auto-increment — MySQL AUTO_INCREMENT, PostgreSQL SERIAL/BIGSERIAL, SQLite inline INTEGER PRIMARY KEY AUTOINCREMENT, SQL Server IDENTITY(1,1).

Is my data uploaded anywhere?

No. The tool runs entirely in your browser — your schema never leaves your machine.

Can I generate DDL for a dialect not listed?

The four supported dialects cover MySQL, PostgreSQL, SQLite, and SQL Server. Output for a similar engine (e.g. MariaDB) is usually compatible with the MySQL setting.