PostgreSQL syntax by default. Toggle for MySQL, SQLite, SQL Server. Most differences are in quoting and some type names.

Yes — from the first few rows. Columns that look integer stay integer; mixed data becomes text. Override per column if detection is wrong.

Bulk inserts or one per row?

Configurable — single INSERT with many rows (faster), or one INSERT per row (more readable). 1000 rows per batch is a common sweet spot.

What about auto-generated IDs?

Skip the ID column in the CSV; the database handles it. Or include it if you want specific IDs for reproducibility.

Generate SQL INSERT statements from CSV data. Part of the DevTools Surf developer suite. Browse more tools in the Data / SQL collection.

The SQL INSERT statement was part of the original SQL specification published by IBM in 1974, making it over 50 years old.
CSV format was first used on IBM Fortran systems in 1972, predating personal computers by over a decade.
A single PostgreSQL INSERT with a VALUES list can handle up to 1,000 rows before some drivers start chunking automatically.

Which SQL dialect?: PostgreSQL syntax by default. Toggle for MySQL, SQLite, SQL Server. Most differences are in quoting and some type names.
Does it infer types?: Yes — from the first few rows. Columns that look integer stay integer; mixed data becomes text. Override per column if detection is wrong.
Bulk inserts or one per row?: Configurable — single INSERT with many rows (faster), or one INSERT per row (more readable). 1000 rows per batch is a common sweet spot.
What about auto-generated IDs?: Skip the ID column in the CSV; the database handles it. Or include it if you want specific IDs for reproducibility.

Generate SQL INSERT statements from CSV data. Part of the DevTools Surf developer suite. Browse more tools in the Data / SQL collection.

The SQL INSERT statement was part of the original SQL specification published by IBM in 1974, making it over 50 years old.
CSV format was first used on IBM Fortran systems in 1972, predating personal computers by over a decade.
A single PostgreSQL INSERT with a VALUES list can handle up to 1,000 rows before some drivers start chunking automatically.

Which SQL dialect?: PostgreSQL syntax by default. Toggle for MySQL, SQLite, SQL Server. Most differences are in quoting and some type names.
Does it infer types?: Yes — from the first few rows. Columns that look integer stay integer; mixed data becomes text. Override per column if detection is wrong.
Bulk inserts or one per row?: Configurable — single INSERT with many rows (faster), or one INSERT per row (more readable). 1000 rows per batch is a common sweet spot.
What about auto-generated IDs?: Skip the ID column in the CSV; the database handles it. Or include it if you want specific IDs for reproducibility.