---
title: Introduction
description: Prisma-inspired TypeScript ORM without code generation or WASM query-engine deployment issues
icon: database
---

VibORM is a **Prisma-inspired TypeScript ORM with zero code generation**: types are inferred from your schema the moment you save, and there is no WASM query engine, so it runs anywhere TypeScript runs — including edge runtimes like Cloudflare Workers.

:::info[Status]
VibORM 1.x is the stable release line. Releases follow semantic versioning; every release is listed in the [changelog](https://github.com/beynar/viborm/blob/main/CHANGELOG.md). The project has a single maintainer. Security fixes ship for the latest stable major, as the [security policy](https://github.com/beynar/viborm/security/policy) describes. Coming from 0.1 or a release candidate, read [Upgrading to V1](/docs/getting-started/upgrading-to-v1) first.
:::

## What It Looks Like

```typescript
import { s } from "viborm";
import { createClient } from "viborm/pg";

// Define schema - types are inferred from this
const user = s.model({
  id: s.string().id().ulid(),
  email: s.string().unique(),
  name: s.string().nullable(),
  posts: s.toMany(() => post),
});

const post = s.model({
  id: s.string().id().ulid(),
  title: s.string(),
  published: s.boolean().default(false),
  authorId: s.string(),
  author: s.toOne(() => user).fields("authorId").references("id"),
});

// Create client
const client = createClient({
  schema: { user, post },
  databaseUrl: process.env.DATABASE_URL,
});

// Prisma-inspired queries, fully typed
const users = await client.user.findMany({
  where: { email: { contains: "@company.com" } },
  include: { posts: { where: { published: true } } },
});
```

If you know Prisma, this should feel familiar: `findMany`, `findFirst`, `create`, `update`, `delete`, nested `include`, and `where` filters follow the same patterns. The public contract is intentionally narrower than Prisma today — see [Compatibility](/docs/client/compatibility) for supported operations and known differences.

## Features

- **PostgreSQL, MySQL, SQLite, and SurrealDB** supported, SurrealDB in its own SurrealQL
- **Implicit many-to-many** — define relations without manual join tables
- **Rich filtering** — `contains`, `startsWith`, `in`, `not`, and more
- **Deep relation filtering** — `where: { posts: { some: { published: true } } }`
- **Single query execution** — relations loaded in one roundtrip, no N+1
- **Nested writes** — create or connect related records in one operation where the [compatibility contract](/docs/client/compatibility) lists support
- **Standard Schema integration** — extend any scalar with Zod, Valibot, or any [Standard Schema](https://standardschema.dev/) validator via `.schema()`

Beyond schema definition, every operation schema (where, create, update) is Standard Schema V1 compliant and exportable to JSON Schema — reuse them in API routes, OpenAPI generation, or AI tool calling via `getSchemas`. See [Standard Schema](/docs/internals/validation/standard-schema).

## Next Steps

**[Install and Choose a Driver](/docs/getting-started)**

Install VibORM and select the right database client

**[Quick Start](/docs/getting-started/quick-start)**

Go from schema definition to your first typed query

**[Compatibility](/docs/client/compatibility)**

Check supported operations and database differences

## Why VibORM?

VibORM started as an experiment in how far vibe coding can go, driven by two Prisma friction points: the `prisma generate` step and WASM query-engine deployment issues. Read [the full story](/docs/why).
