> For the complete documentation index, see [llms.txt](https://docs.agenticflow.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.agenticflow.ai/integrations/connection-providers/media/kling.md).

# Kling

Use your Kling account to generate images, video, and audio, create avatars and lip-sync clips, prepare video edits, and produce commerce media.

## Create the connection

1. Get your credentials from your Kling API account. The [Kling Open Platform](https://kling.ai/document-api/api/get-started/authentication) provides account authentication guidance.
2. In AgenticFlow, open **Connections**, choose **Kling**, and select **Create Connection**.
3. Give the connection a recognizable name.
4. Choose the authentication mode that matches your credentials:
   * **API key:** enter your **API Key**.
   * **Legacy JWT:** enter your **Legacy Access Key** and **Legacy Secret Key**. AgenticFlow handles authentication; you do not need to construct a token.
5. Save, then select this connection in a Kling node.

These nodes use Kling’s international API service. Use credentials issued for that service. Model and feature access depend on your Kling account.

## Available nodes

| Node                                                                    | What you can do                                                                                                       |
| ----------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------- |
| [Generate Image](/reference/nodes/kling_group_image_generation.md)      | Generate images, combine image references, complete a subject, or extend an image beyond its original edges.          |
| [Generate Video](/reference/nodes/kling_group_video_generation.md)      | Generate videos from text or images, guide motion with references, use Omni models, or create supported effects.      |
| [Generate Audio](/reference/nodes/kling_group_audio_generation.md)      | Generate speech or sound from text, or create audio for a video.                                                      |
| [Avatar and Lip Sync](/reference/nodes/kling_group_avatar_lipsync.md)   | Create avatar videos, synchronize lips with audio, and inspect the face or image inputs needed for supported actions. |
| [Edit Video Selections](/reference/nodes/kling_group_video_editing.md)  | Prepare selected areas of a video for a multi-element edit before submitting the final editing task.                  |
| [Commerce Media](/reference/nodes/kling_group_commerce.md)              | Create commerce media for apparel, product scenes, product videos, or virtual try-on.                                 |
| [Manage Voices and Elements](/reference/nodes/kling_group_resources.md) | Create and manage reusable visual elements and custom voices, or browse provider presets.                             |
| [Get and List Tasks](/reference/nodes/kling_group_tasks.md)             | Check an existing Kling task or list tasks so later workflow steps can use completed media.                           |
| [Usage and Billing](/reference/nodes/kling_group_account.md)            | Review your connected Kling account’s resource packages and usage records.                                            |

Open a node reference for its inputs, outputs, example, and full list of **Model / Action** choices. Individual options depend on provider availability.

## Your first image-to-video workflow

1. Add **Kling: Generate Video** and select your Kling connection.
2. Choose **Kling 3.0 Turbo — Image To Video** under **Model / Action**.
3. Provide an image and describe the movement you want, such as a slow camera move around a product.
4. Complete the duration and other settings offered by the form, then save and run.
5. Keep the returned task ID.
6. Add **Kling: Get and List Tasks**, select **Get Tasks by ID — Video**, and use that ID.
7. If the task is still processing, check the same task later. Continue to a media-consuming step only after the result is ready.

## Choose the matching result action

Kling actions can create different kinds of tasks. Use the query for the task you created: commerce tasks use commerce queries; image, audio, avatar, lip-sync, and older task types have their own query choices.

**Edit Video Selections** prepares areas for a multi-element edit. After reviewing the selection preview, use the multi-elements creation action in **Generate Video** to submit the final edit.

Reusable voices and elements can be selected or managed through **Manage Voices and Elements**. Check a custom resource’s readiness before using it in generation.

## Results and common questions

* **I see a task ID but no video:** generation was submitted and may still be processing. Use the matching task query.
* **The task cannot be found:** check the task type and use the same connection that created it.
* **A model rejects my image or video:** review the required file type, dimensions, duration, and references for the selected model.
* **I already use Kling in Higgsfield:** that uses your Higgsfield account. This connection uses your Kling account and its allowance separately.

See [model-specific forms](/workflows/media-action-forms.md) and [media results and AI Drive](/workflows/workflow-single-run/media-results.md).