> For the complete documentation index, see [llms.txt](https://docs.stackai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stackai.com/integrations/catalog/qdrant.md).

# Qdrant

Connect to Qdrant to create vector collections, upsert and delete points, and retrieve records by vector similarity or payload filters.

## Connectors

Choose the connector that matches your authentication method. The options below use the names shown in StackAI.

| Connector | Authentication   |
| --------- | ---------------- |
| Qdrant    | Database API Key |

### Qdrant

Connect to a Qdrant Cloud or self-hosted vector database to create collections, write points, run vector and payload searches, and delete points.

#### Set up in Qdrant

1. Prepare the Qdrant account or service that the workflow will access. Grant access to the required resources and operations.
2. **Deployment URL:** The HTTPS Qdrant database REST endpoint, including the port when needed. Use the cluster endpoint from Qdrant Cloud or the TLS-enabled base URL of a self-hosted deployment, with no path.
3. **Database API Key:** A Qdrant database API key sent in the api-key header. Use manage access to create collections, collection read-write access for point writes, or collection read access for search and scroll.

#### Set up in StackAI

1. Open **Settings → Connections**, add a connection, and select **Qdrant**.
2. Enter the credentials and connection details in the fields below.
3. Save the connection and check its connection status. Select this saved connection when configuring an action or feature that uses it.

**Connection fields**

| StackAI field    | Required | What to enter                                                                                                                                                                                                                                                                    |
| ---------------- | -------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Deployment URL   | Yes      | The HTTPS Qdrant database REST endpoint, including the port when needed. Use the cluster endpoint from Qdrant Cloud or the TLS-enabled base URL of a self-hosted deployment, with no path. [Provider instructions](https://qdrant.tech/documentation/cloud/cluster-access/).     |
| Database API Key | Yes      | A Qdrant database API key sent in the api-key header. Use manage access to create collections, collection read-write access for point writes, or collection read access for search and scroll. [Provider instructions](https://qdrant.tech/documentation/cloud/authentication/). |

### Verify the connection

Check that the connection is healthy. Then run a read operation against a resource the connected account can access. A successful connection check confirms authentication; it does not guarantee permission for every action.

If access is denied, check the credential or OAuth permissions, the selected account or environment, and access to the specific resource. Update expired credentials or reconnect after permissions change.

## Use an action

1. Add the required **Qdrant** action to the workflow.
2. Select a connection when the action requests one.
3. Complete the required fields. Use the reference picker when a value comes from another workflow step.
4. Run the workflow and inspect the action’s output. Check the result before using it in a later step.

## Actions

### Create Collection

Create a Qdrant collection for a new set of embedding vectors.

#### Inputs

| Field           | Type            | Required | Description                                                                                                          |
| --------------- | --------------- | -------- | -------------------------------------------------------------------------------------------------------------------- |
| Collection name | string          | Yes      | Name for the new vector collection.                                                                                  |
| Vector size     | integer         | Yes      | Number of dimensions in every vector stored in this collection.                                                      |
| Distance metric | select (string) | Yes      | Method Qdrant uses to compare vectors. Available values: Cosine, Euclidean, Dot product, Manhattan. Default: Cosine. |
| Vector name     | string          | No       | Optional name for this vector configuration.                                                                         |

#### Outputs

| Field           | Type    | Description                               |
| --------------- | ------- | ----------------------------------------- |
| Collection name | string  | Name of the created collection.           |
| Created         | boolean | Whether Qdrant created the collection.    |
| Elapsed seconds | number  | Time Qdrant spent processing the request. |

### Delete Points

Delete points by id or by a Qdrant payload filter.

#### Inputs

| Field                | Type            | Required | Description                                                                                               |
| -------------------- | --------------- | -------- | --------------------------------------------------------------------------------------------------------- |
| Collection           | select (string) | Yes      | Collection from which points will be deleted. Options load from the selected connection.                  |
| Delete by            | select (string) | Yes      | Choose explicit point ids or a payload filter. Available values: Point ids, Payload filter. Default: ids. |
| Point ids            | string          | No       | Comma-separated uint64 or UUID ids to delete.                                                             |
| Payload filter       | textarea        | No       | Qdrant Filter JSON selecting every point to delete.                                                       |
| Show advanced fields | boolean         | No       | Reveal completion, ordering, and timeout controls. Default: false.                                        |
| Wait for completion  | boolean         | No       | Wait until the point deletions are applied. Default: true.                                                |
| Write ordering       | select (string) | No       | Consistency guarantee for ordering this deletion. Available values: Weak, Medium, Strong. Default: weak.  |
| Timeout in seconds   | integer         | No       | Optional provider-side timeout for the deletion. Minimum: 1.                                              |

#### Outputs

| Field           | Type    | Description                                               |
| --------------- | ------- | --------------------------------------------------------- |
| Status          | string  | Qdrant deletion status.                                   |
| Operation id    | integer | Provider sequence number for the deletion when available. |
| Elapsed seconds | number  | Time Qdrant spent processing the request.                 |

### Filtered Search

Scroll through points that match a Qdrant payload filter.

#### Inputs

| Field                 | Type            | Required | Description                                                                          |
| --------------------- | --------------- | -------- | ------------------------------------------------------------------------------------ |
| Collection            | select (string) | Yes      | Collection whose points will be scanned. Options load from the selected connection.  |
| Payload filter        | textarea        | Yes      | Qdrant Filter JSON that selects the points to return.                                |
| Maximum results       | integer         | No       | Soft cap on matching points returned across fetched pages. Minimum: 1. Default: 100. |
| Include payload       | boolean         | No       | Return point metadata with each record. Default: true.                               |
| Show advanced options | boolean         | No       | Reveal paging, vector-output, and timeout controls. Default: false.                  |
| Page size             | integer         | No       | Points requested from Qdrant per scroll page. Minimum: 1. Default: 10.               |
| Starting point id     | string          | No       | Point id cursor from which scrolling resumes.                                        |
| Include vectors       | boolean         | No       | Return stored vectors with each record. Default: false.                              |
| Timeout in seconds    | integer         | No       | Optional provider-side timeout for each scroll request. Minimum: 1.                  |

#### Outputs

| Field            | Type          | Description                                                       |
| ---------------- | ------------- | ----------------------------------------------------------------- |
| Points           | object\_array | Typed list of matching points with payloads and optional vectors. |
| Next page offset | string        | Point id cursor for the next scroll page.                         |
| Elapsed seconds  | number        | Total provider-reported processing time across fetched pages.     |

### Search Points

Search points by vector similarity for retrieval and RAG workflows.

#### Inputs

| Field                 | Type            | Required | Description                                                                                                         |
| --------------------- | --------------- | -------- | ------------------------------------------------------------------------------------------------------------------- |
| Collection            | select (string) | Yes      | Collection to search. Options load from the selected connection.                                                    |
| Query vector          | string          | Yes      | Embedding to compare against stored vectors, as a JSON array or comma-separated numbers.                            |
| Vector name           | select (string) | No       | Named vector to search. Required when the collection uses named vectors. Options load from the selected connection. |
| Maximum results       | integer         | No       | Soft cap on scored points returned across exact-search pages. Minimum: 1. Default: 10.                              |
| Include payload       | boolean         | No       | Return point metadata with each match. Default: true.                                                               |
| Show advanced options | boolean         | No       | Reveal filter, paging, threshold, and response controls. Default: false.                                            |
| Payload filter        | textarea        | No       | Optional Qdrant Filter JSON that narrows matching points.                                                           |
| Page size             | integer         | No       | Points requested from Qdrant per exact-search page. Minimum: 1. Default: 10.                                        |
| Starting offset       | integer         | No       | Number of ranked results to skip before returning points. Minimum: 0. Default: 0.                                   |
| Score threshold       | number          | No       | Only return points whose score passes this threshold.                                                               |
| Include vectors       | boolean         | No       | Return stored vectors with each match. Default: false.                                                              |
| Timeout in seconds    | integer         | No       | Optional provider-side timeout for the search. Minimum: 1.                                                          |

#### Outputs

| Field           | Type          | Description                                                            |
| --------------- | ------------- | ---------------------------------------------------------------------- |
| Scored points   | object\_array | Typed list of matching points, scores, payloads, and optional vectors. |
| Next offset     | integer       | Offset to resume the exact ranked results when more may be available.  |
| Elapsed seconds | number        | Total provider-reported processing time across fetched pages.          |

### Upsert Points

Insert or replace points with vectors and optional payloads.

#### Inputs

| Field                | Type            | Required | Description                                                                                                                        |
| -------------------- | --------------- | -------- | ---------------------------------------------------------------------------------------------------------------------------------- |
| Collection           | select (string) | Yes      | Collection that will receive the points. Options load from the selected connection.                                                |
| Points               | textarea        | Yes      | JSON array of point ids, vectors, and optional payload metadata.                                                                   |
| Show advanced fields | boolean         | No       | Reveal write mode, completion, ordering, and timeout controls. Default: false.                                                     |
| Update mode          | select (string) | No       | Choose whether existing and missing point ids may be written. Available values: Upsert, Insert only, Update only. Default: upsert. |
| Wait for completion  | boolean         | No       | Wait until the point changes are applied. Default: true.                                                                           |
| Write ordering       | select (string) | No       | Consistency guarantee for ordering this write. Available values: Weak, Medium, Strong. Default: weak.                              |
| Timeout in seconds   | integer         | No       | Optional provider-side timeout for the write. Minimum: 1.                                                                          |

#### Outputs

| Field           | Type    | Description                                            |
| --------------- | ------- | ------------------------------------------------------ |
| Status          | string  | Qdrant update status.                                  |
| Operation id    | integer | Provider sequence number for the write when available. |
| Elapsed seconds | number  | Time Qdrant spent processing the request.              |

## Important behavior

Actions that create, update, send, or delete data can change the connected service. Review their inputs before running them. A retry can repeat an operation; consider this when configuring retries for actions that change data.

The catalog lists the fields and capabilities available in the supplied provider definition. Availability can also depend on your deployment and organization’s Feature Access settings.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.stackai.com/integrations/catalog/qdrant.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
