This minimal template demonstrates how to build semantic image search with natural language queries using the Mixedbread Search API.

• Bun (or Node.js 22+)
• Git

git clone <your-repo-url>
cd mxbai-image-search

bun install

Create a .env file in the root directory:

cp .env.example .env

Add your Mixedbread credentials to the .env file:

MXBAI_API_KEY=your-api-key-here
MXBAI_STORE_ID=your-store-id

To get your API key and Store ID, you have two options:

1. From Vercel Integration (Recommended if deploying to Vercel):

   • Go to your Vercel Dashboard
   • Navigate to your project's Integrations tab
   • Install or access the Mixedbread integration
   • Copy your API key and Store ID from the integration settings

2. From Mixedbread Platform (For standalone use):

   • Visit the Mixedbread Platform
   • Sign up or log in to your account
   • Navigate to API Keys and create a new key
   • Navigate to Stores and create a new Store, then copy the Store ID

The project includes a script to upload images to your Mixedbread Store:

bun scripts/upload-images.ts ./public/images

This will process all images in the ./public/images directory and upload them to your Store.

Start the development server:

bun dev

Open http://localhost:3000 in your browser and start searching for images using natural language queries.

• Mixedbread Documentation - Learn about Mixedbread's features and APIs
• Quickstart Guide - Get started with creating Stores and uploading files
• Mixedbread Discord - Join the community and get support