SQL to ERD Diagram Generator

Upload a MySQL .sql file and instantly get an interactive entity-relationship diagram. The parser reads CREATE TABLE and ALTER TABLE statements to extract tables, columns, primary keys, foreign keys, unique keys and indexes — then draws them on a zoomable, draggable canvas. Everything runs in your browser, so your schema never leaves your device. Full guide →

Read a schema fast with an ERD built from SQL

What is this tool?

Paste your `CREATE TABLE` statements (MySQL, PostgreSQL, SQLite…) and the tool draws an entity-relationship diagram: each table is a node, each foreign key is an edge. Drag, arrange, and see how tables relate — no database install or desktop tool required.

Everything runs in your browser; the SQL you paste never leaves the page. Handy for internal or sensitive schemas you still want to visualise and share in docs or reviews.

When should you use it?

When inheriting an unfamiliar codebase: paste the schema dump to grasp which tables link to which before writing a query or migration. When designing a feature: sketch a few tables and sanity-check the FK relationships.

When writing technical docs or prepping a review: export a compact diagram instead of describing relationships in prose. When onboarding teammates: an ERD image lands far faster than hundreds of lines of DDL.

Supported SQL syntax

It focuses on `CREATE TABLE` with column names, types, `PRIMARY KEY`, and relationships via `FOREIGN KEY ... REFERENCES other_table(col)`. Inline column references (e.g. `user_id INT REFERENCES users(id)`) are detected too.

Tip: end each `CREATE TABLE` with `;` and place referenced tables in any order — the parser links by table name, not position. Non-structural statements (INSERT, comments) can be omitted to keep the diagram clean.

Quick usage example

Paste two tables: `users(id PK, email)` and `orders(id PK, user_id FK → users.id, total)`. The diagram shows two nodes and an edge from orders → users, confirming the one-to-many relationship at a glance.

Need to share it? Arrange the nodes, screenshot, and drop it into a PR or design doc — readers grasp the structure in seconds instead of parsing DDL line by line.

Common parse errors

A missing `;` between tables, or pasted `ENGINE=...`/`AUTO_INCREMENT` noise, can make a table go undetected. If a node is missing, isolate that table and paste it alone to pinpoint the offending statement.

A relationship edge that does not appear usually means the table/column in `REFERENCES` does not match exactly (wrong case, a stray schema prefix like `public.`). Normalise the names to match the target table and the edge connects.

Related Tools

All tools

QR Code & Barcode Generator

Generate customizable QR codes with colors, logos, and download options

CSV to Excel

Convert CSV files to Excel format (XLSX or XLS) with auto-sized columns

Markdown to Word

Convert Markdown (.md) files to Word documents (.docx) with formatting

Vector Embeddings Formatter

Format vector embeddings into JSON, Python, SQL (pgvector), Pinecone, or MongoDB syntax

About SQL to ERD: Visualize Your Database Schema

Paste or upload SQL DDL and turn it into a clean entity-relationship diagram. Supports MySQL CREATE TABLE / ALTER TABLE with primary keys, composite keys, foreign keys, unique keys, indexes, defaults, AUTO_INCREMENT and comments. Auto-layout with Dagre, then export to JSON, PNG or SVG.

Key Features

-Parse MySQL CREATE TABLE and ALTER TABLE DDL
-Detect primary keys, foreign keys, unique keys and indexes
-Interactive canvas — zoom, pan, drag tables, minimap
-Automatic graph layout (Dagre) with one-click re-layout
-Searchable sidebar to jump to any table
-Export the parsed model as JSON, or the diagram as PNG/SVG
-100% client-side — your schema never leaves the browser

Use Cases

Understand a Legacy Database

Drop in a production schema dump and see every table and relationship at a glance before making changes.

Documentation & Onboarding

Export a PNG or SVG of the ER diagram for READMEs, wikis, and onboarding docs so new engineers ramp up faster.

Schema Reviews

Visualize migrations and pull-request schema changes to catch missing foreign keys or denormalized tables.

Data Modeling

Sketch and validate relationships while designing a new schema, then export the structured JSON model.

Frequently Asked Questions

Which SQL dialects are supported?▾

MySQL DDL (CREATE TABLE and ALTER TABLE … ADD) is fully supported today, including backticked identifiers, schema prefixes, composite keys and foreign keys. PostgreSQL is parsed on a best-effort basis and full support is planned.

Is my SQL uploaded anywhere?▾

No. Parsing and rendering happen entirely in your browser. Your .sql file and schema never leave your device.

What does the tool extract?▾

Tables, columns with types, nullability, defaults and comments, AUTO_INCREMENT flags, primary keys (including composite), unique keys, indexes, and foreign-key relationships from both CREATE TABLE constraints and ALTER TABLE statements.

Can I export the diagram?▾

Yes. Export the full parsed structure as JSON, or export the rendered diagram as a PNG or SVG image.

What happens with SQL it cannot parse?▾

Unsupported statements are skipped and listed in a warnings panel, so a single unusual statement never breaks the rest of the diagram.

Tips & Best Practices

-Use Auto Layout after loading to arrange tables left-to-right by relationship
-Click a table in the sidebar to center the canvas on it
-Export to SVG for crisp, scalable diagrams in documentation
-Load the bundled sample to see how the output looks before uploading your own schema