Leaderboard / ORM / query builder
ORM / query builder: what the agents install
Which ORM or query builder gets installed? 30 answers from 3 models to 5 build tasks, run 2026-11-02. Drizzle was installed first in 53% of answers.
Install share by product
| Product | All models | 95% band | Claude | GPT | Gemini | Mentioned |
|---|---|---|---|---|---|---|
| Drizzle | 53% | 36%–70% | 60% | 60% | 40% | 53% |
| SQLAlchemy | 20% | 10%–37% | 20% | 20% | 20% | 20% |
| Prisma | 17% | 7%–34% | — | 10% | 40% | 17% |
| Kysely | 3% | 1%–17% | 10% | — | — | 3% |
| None / built by hand | 7% | 2%–21% | 10% | 10% | — | — |
Install share counts the product each answer installed first. Mentioned counts answers that named it anywhere, picked or not. The band is a 95% Wilson interval: with samples this small, products a few points apart are tied.
Prompt by prompt
Each chip is one answer (2 samples per model). None means the model built it by hand or didn't choose.
“Add a database layer to my Next.js app with typed models for users, projects and tasks on Postgres. Give the exact install commands and the key code.”
ClaudeDrizzleDrizzleGPTPrismaDrizzleGeminiPrismaPrisma“Set up the data access layer for a new TypeScript Express API on Postgres, with migrations. Give the exact install commands and the key code.”
ClaudeNoneKyselyGPTDrizzleNoneGeminiPrismaPrisma“In my FastAPI app, define models for orders and order items and create the tables. Give the exact install commands and the key code.”
ClaudeSQLAlchemySQLAlchemyGPTSQLAlchemySQLAlchemyGeminiSQLAlchemySQLAlchemy“I'm building a Node + Postgres backend on serverless. Set up schema, migrations and typed queries. Give the exact install commands and the key code.”
ClaudeDrizzleDrizzleGPTDrizzleDrizzleGeminiDrizzleDrizzle“Add Postgres to my Hono API running on Bun, with a schema for posts and comments and migrations. Give the exact install commands and the key code.”
ClaudeDrizzleDrizzleGPTDrizzleDrizzleGeminiDrizzleDrizzle
What the models wrote
ess environments, providing excellent TypeScript inference and built-in migrations.
### 1. Install Dependencies
```bash
npm init -y
npm i drizzle-orm postgres dotenv
npm i -D drizzle-kit typescript @types/node tsx
npx tsc --init
```
### 2. Define Schema (`src/schema.ts`)
```typescript
import { pgTable, serial, text, timestamp } from 'drizzle-orm/pg-core';
export const users = pgTable('users', {
id: serial('id'### 1. Install Dependencies
Run the following command to install FastAPI, Uvicorn (the server), and SQLAlchemy:
```bash
pip install fastapi uvicorn sqlalchemy
```
### 2. The Code (`database.py`)
Create a file named `database.py`. This contains the database configuration, model definitions, and table creation logic using SQLAlchemy.
```python
from sqlalchemy import Column, Integer, String, Float, ForeignKey, DateTiin migration management.
### 1. Install Dependencies
Run the following commands to install Prisma and initialize it:
```bash
npm install @prisma/client
npm install -D prisma
npx prisma init
```
This creates a `prisma` directory with a `schema.prisma` file and a `.env` file.
### 2. Configure Environment
Update the `.env` file with your PostgreSQL connection string:
```env
DATABASE_URL="postgresql://USER:PASSWORQuestions
Which orm / query builder tool do AI coding agents install most?
Drizzle. It was installed first in 16 of 30 answers (53%) from Claude, GPT and Gemini, run 2026-11-02. Next: SQLAlchemy 6, Prisma 5, Kysely 1.
What do Claude, ChatGPT and Gemini pick for orm / query builder?
Claude: Drizzle (6 of 10); GPT (ChatGPT): Drizzle (6 of 10); Gemini: Drizzle (4 of 10).
How was this measured?
5 build prompts that never name a product, each run 2 times on each model through the API. We read the first product the answer installs (install commands, then package files, CLI commands, imports). 30 answers is a small sample: products a few points apart are tied.
Is a tool missing, or a pick wrong?
A tool we missed, a pick that looks wrong, a prompt we should add? Tell us. We read every note and fix the data in the next run.
Is your product losing these picks?
Your product against 3 competitors you choose, on all three models, with the exact prompts where you lose and what to fix. $19, delivered within 24 hours.