QyverLabs Docs
  • 👋Welcome
  • Getting Started
    • 👀Why Qyver?
    • 💻Setup Qyver
    • 🏗️Basic Building Blocks
  • Run In Production
    • 🎯Overview
    • 💻Setup Qyver Server
      • 🧮Configuring your app
      • 📔Interacting with app via API
    • ⚙️Supported Vector Databases
      • Redis
      • Mongo DB
      • Qdrant
  • Concepts
    • 🗓️Combining Multiple Embeddings for Better Retrieval Outcomes
    • 🏹Dynamic Parameters/Query Time weights
  • Reference
    • 🎯Overview
    • ⏰Changelog
  • Help & FAQ
    • 📒Logging
    • ❔Support
Powered by GitBook
On this page
  1. Run In Production
  2. Supported Vector Databases

Qdrant

Integrating Qdrant with Qyver

This guide explains how to configure and integrate Qdrant with Qyver.

Configuring an Existing Managed Qdrant Instance

To use Qdrant with Qyver, you can start a managed Qdrant instance (a free tier is available). Instructions for setting up a managed instance are included in the Starting a Managed Qdrant Instance section below.

Once the instance is running, ensure:

  • The server can access the Qdrant instance.

  • Authentication settings are correctly configured.


Configuration Modifications

To integrate Qdrant, add the QdrantVectorDatabase class and include it in the executor:

from qyver import framework as qv

vector_database = qv.QdrantVectorDatabase(
    "<your_qdrant_url>",  # (Required) Qdrant server URL (typically includes a port)
    "<your_api_key>",  # (Required) API key for authentication
    default_query_limit=10,  # (Optional) Maximum number of results per query (defaults to 10)
    # Additional parameters can be passed as kwargs following Qdrant's official Python client documentation.
)

Once configured, assign it to the executor:

...
executor = qv.RestExecutor(
    sources=[source],
    indices=[index],
    queries=[qv.RestQuery(qv.RestDescriptor("query"), query)],
    vector_database=vector_database,  # Use the QdrantVectorDatabase instance
)
...

Starting a Managed Qdrant Instance

To create a managed Qdrant instance, follow these steps:

  1. Click Overview in the left menu.

  2. Select Create Cluster (free-tier or production-ready).

    • Free-tier includes 0.5 vCPU, 1GB RAM, 4GB storage on one node.

    • Paid plans offer customization options like platform selection, region, and high availability (HA).

  3. Once the cluster is created, generate an API key and store it securely (it cannot be retrieved later).

  4. Use this API key in your QdrantVectorDatabase configuration.

PreviousMongo DBNextCombining Multiple Embeddings for Better Retrieval Outcomes

Last updated 3 months ago

Sign in to .

⚙️
Qdrant Cloud