---
title: "Archetype AI"
lang: "en-US"
draft: false
description: "Learn about how to set up a VDP Archetype AI component https://github.com/instill-ai/instill-core"
---
The Archetype AI component is an AI component that allows users to connect the AI models served on the Archetype AI Platform.
It can carry out the following tasks:
- [Describe](#describe)
- [Summarize](#summarize)
- [Upload File](#upload-file)
## Release Stage
`Alpha`
## Configuration
The component configuration is defined and maintained [here](https://github.com/instill-ai/component/blob/main/ai/archetypeai/v0/config/definition.json).
## Setup
In order to communicate with Archetype AI, the following connection details need to be
provided. You may specify them directly in a pipeline recipe as key-value pairs
withing the component's `setup` block, or you can create a **Connection** from
the [**Integration Settings**](https://www.instill.tech/docs/vdp/integration)
page and reference the whole `setup` as `setup:
${connection.<my-connection-id>}`.
| Field | Field ID | Type | Note |
| :--- | :--- | :--- | :--- |
| API Key (required) | `api-key` | string | Fill in your Archetype AI API key |
## Supported Tasks
### Describe
Describe a video.
| Input | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| Task ID (required) | `task` | string | `TASK_DESCRIBE` |
| Query (required) | `query` | string | A guide to describe the video |
| File IDs (required) | `file-ids` | array[string] | The IDs of the videos to describe. These must have been previously uploaded via TASK_UPLOAD_FILE. |
| Output | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| [Descriptions](#describe-descriptions) | `descriptions` | array[object] | A set of descriptions corresponding to different moments in the video |
<details>
<summary> Output Objects in Describe</summary>
<h4 id="describe-descriptions">Descriptions</h4>
| Field | Field ID | Type | Note |
| :--- | :--- | :--- | :--- |
| Description | `description` | string | The description of the frame |
| Frame ID | `frame-id` | integer | The frame number in the video that is being described |
| Timestamp | `timestamp` | number | The moment of the video (in seconds since the start) that is being described |
</details>
### Summarize
Summarize the image.
| Input | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| Task ID (required) | `task` | string | `TASK_SUMMARIZE` |
| Query (required) | `query` | string | A guide to summarize the image |
| File IDs (required) | `file-ids` | array[string] | The IDs of the images to summarize. These must have been previously uploaded via TASK_UPLOAD_FILE. |
| Output | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| Response | `response` | string | A text responding to the query |
### Upload File
Upload file.
| Input | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| Task ID (required) | `task` | string | `TASK_UPLOAD_FILE` |
| File (required) | `file` | string | The file to upload. Accepted formats are JPEG and PNG for images or MP4 for videos |
| Output | ID | Type | Description |
| :--- | :--- | :--- | :--- |
| File ID | `file-id` | string | The ID to reference the file in queries |