Paste a CSV sample, get a T-SQL-dialect DDL with NVARCHAR, DATETIME2, and UNIQUEIDENTIFIER types.
Open the Schema Inferrer →Generate a SQL Server CREATE TABLE from any CSV sample, with type mappings that match T-SQL conventions: BIT for booleans, UNIQUEIDENTIFIER for UUIDs, DATETIME2 for high-precision timestamps, and NVARCHAR(N) for Unicode-safe string columns.
Sample values and the dialect-native type they map to.
| Sample value | Inferred type |
|---|---|
| 1, 1000 | INT |
| 12345678901234 | BIGINT |
| 12.50 | DECIMAL(10,2) |
| true, false, 0, 1 | BIT |
| 2024-03-15 | DATE |
| 2024-03-15 09:30:00.123 | DATETIME2(3) |
| a3bb189e-… | UNIQUEIDENTIFIER |
| "Acme Corp" | NVARCHAR(64) |
| paragraph > 4000 chars | NVARCHAR(MAX) |
ticket_id,subject,priority,opened_at,is_resolved 1,"Login broken",2,2024-03-15 09:30:00,0 2,"Cannot upload",1,2024-03-15 10:01:22,1
CREATE TABLE tickets (
ticket_id INT NOT NULL,
subject NVARCHAR(64) NOT NULL,
priority INT NOT NULL,
opened_at DATETIME2 NOT NULL,
is_resolved BIT NOT NULL
);
No signup, no install — runs entirely in your browser.
Open the Schema Inferrer →