What is Inverse Model?

Inverse Model is a NestJS module that turns an existing database into a GraphQL API. You give it a connection string; on start it reads every table or collection, with its columns, keys and foreign keys, and serves a GraphQL schema to read and write all of them.

@Module({
  imports: [InverseModelModule.forRoot({ url: process.env.DATABASE_URL! })],
})
export class AppModule {}

It is database-first: the database is the source of truth, and the API follows it. There are no entities, DTOs or resolvers to write, and nothing is ever migrated or synchronized: the library only reads the schema.

When to use it

• Internal tools and admin panels that need to read and edit data that already exists.
• Modernizing a legacy database behind a typed API without rewriting it.
• Prototypes that need a working backend in minutes.
• A data API inside an existing NestJS app, next to your own modules, guards and resolvers.

What it generates

For every table or collection:

Queries	A list with filters, ordering and limit/offset, a count, a lookup by primary key, and a Relay-style connection with cursors.
Mutations	Create, update, delete, upsert, and batch create and delete.
Relations	A field on both sides of every foreign key, composite keys included, batched per request.

See The generated API for the full shape.

How it compares

Tools like Hasura or PostGraphile run as a separate server with their own configuration format. Inverse Model is a library that runs inside your NestJS process, configured in TypeScript in forRoot, so it shares your app's lifecycle, dependency injection, logging and deployment. It also covers databases those tools support partially or not at all, such as SQL Server and MongoDB.

Supported databases

Database	URL	Driver
PostgreSQL	postgres://user:pass@host:5432/db	pg
MySQL	mysql://user:pass@host:3306/db	mysql2
MariaDB	mariadb://user:pass@host:3306/db	mysql2
SQL Server	mssql://user:pass@host:1433/db	mssql
MongoDB	mongodb://host:27017/db or mongodb+srv://…	mongodb

Every database is tested in CI on every change, both directly and through the full GraphQL API in Docker. See Databases for what differs between them.