wontonee/laravel-qdrant-sdk

Laravel SDK for Qdrant vector database with embedders like OpenAI and Gemini

1.0.1 2025-05-29 09:07 UTC

Requires

• php: ^8.1
• illuminate/http: ^10.0|^11.0|^12.0
• illuminate/support: ^10.0|^11.0|^12.0

MIT

• Wontonee

README

Laravel Qdrant SDK is a Laravel package that provides a simple, expressive interface for working with the Qdrant vector database. It supports collection management, vector upserts, similarity search, and integration with embedders like OpenAI and Gemini.

Features

• Create, update, delete, and list Qdrant collections
• Upsert and manage vector points
• Perform similarity search and recommendations
• Easily integrate with OpenAI or Gemini for vector embeddings
• Simple trait for adding vector support to Eloquent models

Installation

1. Require the package:

   composer require wontonee/laravel-qdrant-sdk

2. Publish the config:

   php artisan vendor:publish --provider="Wontonee\LarQ\Providers\LarQServiceProvider" --tag=larq-config

3. Set your .env variables:

   LARQ_HOST=http://localhost:6333
   LARQ_API_KEY=
   OPENAI_API_KEY=sk-...
   OPENAI_MODEL=text-embedding-ada-002
   GEMINI_API_KEY=...
   GEMINI_MODEL=models/embedding-001

4. Run Qdrant with persistent storage (recommended for local dev):

   docker run -p 6333:6333 -v $(pwd)/qdrant_storage:/qdrant/storage qdrant/qdrant

Usage

Creating a Collection

use Wontonee\LarQ\Qdrant\Collections\CreateCollection;

$vectorParams = [
    'size' => 3,
    'distance' => 'Cosine',
];
$response = (new CreateCollection())->handle('my_collection', $vectorParams);

Upserting Points

use Wontonee\LarQ\Qdrant\Points\UpsertPoints;

$points = [
    [
        'id' => 1,
        'vector' => [0.1, 0.2, 0.3],
        'payload' => ['label' => 'A'],
    ],
];
$response = (new UpsertPoints())->handle('my_collection', $points);

Scrolling Points

use Wontonee\LarQ\Qdrant\Points\ScrollPoints;
$response = (new ScrollPoints())->handle('my_collection', 10);

Listing All Collections

use Wontonee\LarQ\Qdrant\Collections\ListCollections;
$response = (new ListCollections())->handle();

Using the HasVectors Trait

Add vector support to your Eloquent model:

use Wontonee\LarQ\Traits\HasVectors;

class Product extends Model {
    use HasVectors;
    // Optionally override vectorText(), vectorPayload(), etc.
}

Example Controller Test

See QdrantDemoController for a full example of creating a collection, upserting points, and listing them.

Troubleshooting

• Data disappears after restart: Make sure Qdrant is running with persistent storage (see above).
• Connection issues: Check your LARQ_HOST and API key settings in .env.

License

MIT