> ## Documentation Index
> Fetch the complete documentation index at: https://veryfront.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Qdrant

> Manage Qdrant vector database collections and search or upsert points

## At a glance

* **Availability**: Experimental ([how to enable](/cloud/integrations)).
* **Auth**: API key.
* **Connection**: The key from `QDRANT_API_KEY` is sent as the `api-key` header.
* **Docs**: [https://qdrant.tech/documentation/cloud/authentication/](https://qdrant.tech/documentation/cloud/authentication/)

## Credentials

Set these per environment. See [Connect an integration](/cloud/integrations).

| Variable         | Required | Description                                                                                               |
| ---------------- | -------- | --------------------------------------------------------------------------------------------------------- |
| `QDRANT_API_KEY` | Yes      | Qdrant database API key for your cluster [Docs](https://qdrant.tech/documentation/cloud/authentication/). |

## Setup

1. **Create a Qdrant Cloud account**: Sign up at [https://cloud.qdrant.io](https://cloud.qdrant.io) - the free tier includes a 1GB cluster, enough for testing. Self-hosted Qdrant works too (run it with an api-key configured).
2. **Create a cluster and API key**: Create a cluster, then open its API Keys tab and create a database API key. Copy it immediately - it is only shown once.
3. **Store the key and note your endpoint**: Add QDRANT\_API\_KEY=.... to your .env file. Copy the cluster endpoint hostname from Cluster details - tools take it as the clusterHost parameter without scheme or port (e.g. xyz-example.eu-central-1-0.aws.cloud.qdrant.io); requests go to port 6333.
4. **Verify access**: Run the List Collections tool with your clusterHost.

* The API key is sent in the api-key header; Qdrant Cloud also accepts Authorization: Bearer
* Every tool needs the clusterHost parameter (hostname only; requests target HTTPS port 6333) - there is no global Qdrant API host
* Vectors must match the collection's configured vector size; generate embeddings separately (e.g. with OpenAI or Mistral)

Provider API reference: [https://api.qdrant.tech/api-reference](https://api.qdrant.tech/api-reference)

## Tools

| Tool                | Access | Description                                                          |
| ------------------- | ------ | -------------------------------------------------------------------- |
| List Collections    | Read   | List all collections in the Qdrant cluster                           |
| Get Collection Info | Read   | Get a collection's configuration, vector parameters, and point count |
| Search Points       | Read   | Search a collection for the points nearest to a query vector         |
| Upsert Points       | Write  | Insert or update points (vectors with payloads) in a collection      |

## Example prompts

* List the collections in my Qdrant cluster and show each one's vector size and point count.
* Search my Qdrant collection with an embedding vector and show the top matches with their payloads.