## Create Task
`POST /kling/v1/videos/omni-video`
```bash cURL theme={null}
curl --request POST \
--url https://www.anyfast.ai/kling/v1/videos/omni-video \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"model_name": "kling-o1",
"prompt": "The person in <<>> walks into a rain-soaked neon street with a cinematic tracking shot",
"image_list": [
{"image_url": "https://example.com/character.png"}
],
"mode": "pro",
"aspect_ratio": "16:9",
"duration": "7",
"sound": "off"
}'
```
```json 200 theme={null}
{
"id": "860260753860210752",
"task_id": "860260753860210752",
"object": "video",
"model": "kling-video-o1",
"status": "",
"progress": 0,
"created_at": 1773130665
}
```
## Request headers
Bearer authentication in the form `Bearer YOUR_API_KEY`.
Data exchange format. Use `application/json`.
## Request parameters
Model name. Use `kling-o1`.
Generation or editing instruction, up to 2,500 characters. Reference media with `<<>>`, `<<>>`, and `<<>>`.
Reference images, including optional first and end frames.
Image URL or raw Base64 without the `data:image/...;base64,` prefix.
Frame type: `first_frame` or `end_frame`. An end frame requires a first frame.
Legacy elements. Reference them in list order with `<<>>`, `<<>>`, and so on.
Element ID returned by the Legacy Element Management API.
At most one reference video for source-video editing or for content, motion, style, and camera reference.
Publicly accessible video URL. Base64 video input is not supported.
`base` edits the source video. `feature` uses the source as a generation reference.
Preserve source audio with `yes`, or discard it with `no`.
Output mode. `std` is 720p and `pro` is 1080p. Kling O1 does not support `4k`.
`16:9`, `9:16`, or `1:1`. Required without a first-frame or video reference; not supported for source-video editing.
Duration from 3 to 10 seconds. Ignored when `refer_type` is `base`; output follows the source-video duration.
Kling O1 does not support native audio generation. Use `off`; it is required whenever `video_list` is present.
Platform watermark configuration.
Add the platform watermark.
## Parameter rules
### Multimodal references
* Reference images, elements, and videos with `<<>>`, `<<>>`, and `<<>>`. Numbering starts at 1 and follows array order.
* Kling O1 is single-shot only. It does not support `multi_shot`, `shot_type`, or `multi_prompt`.
* First/end-frame mode accepts at most two images. An end frame requires a first frame, and this mode cannot use elements.
### Image requirements
* JPG, JPEG, or PNG, up to 10 MB per image.
* At least 300 px on each side, with an aspect ratio from 1:2.5 to 2.5:1.
* Public URL or raw Base64 without a data URI prefix.
### Video requirements
* MP4 or MOV, at most one video up to 200 MB.
* At least 3 seconds, dimensions from 720 to 2160 px, 24–60 fps, and aspect ratio from 1:2.5 to 2.5:1.
* `refer_type: "base"` edits the source video. It cannot use first/end frames, and `duration` is ignored.
* `refer_type: "feature"` references content, action, style, or camera movement and can generate a previous or next shot.
* Set `sound` to `off` with video input. Use `keep_original_sound` to preserve or remove the source audio.
### Media counts
| Scenario | Limit |
| ------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
| No reference video, multi-image elements only | Reference images plus element images: at most 7 |
| No reference video, video-character elements only | At most 3 video-character elements |
| No reference video, both element types | At most 3 video-character elements; reference images plus multi-image element images: at most 4 |
| Reference video, multi-image elements only | Reference images plus element images: at most 4 |
| Reference video, video-character elements only | At most 1 video-character element |
## Create response
AnyFast task ID.
Task ID, equal to `id`.
Always `video`.
Model used for the task.
Initial task status.
Initial progress.
Creation Unix timestamp.
## Scenario Invocation Examples
### Text to video
```json theme={null}
{
"model_name": "kling-o1",
"prompt": "A sailboat crosses golden morning fog as an aerial camera slowly moves forward",
"mode": "pro",
"aspect_ratio": "16:9",
"duration": "6",
"sound": "off"
}
```
### Image and element reference
```json theme={null}
{
"model_name": "kling-o1",
"prompt": "The person in <<>> wears the outfit from <<>> and enters a cafe",
"image_list": [
{"image_url": "https://example.com/character.png"}
],
"element_list": [
{"element_id": 123456789}
],
"mode": "pro",
"aspect_ratio": "16:9",
"duration": "5",
"sound": "off"
}
```
### Video editing
```json theme={null}
{
"model_name": "kling-o1",
"prompt": "Change the weather in <<>> to snow and apply a cinematic color grade",
"video_list": [
{
"video_url": "https://example.com/source.mp4",
"refer_type": "base",
"keep_original_sound": "yes"
}
],
"mode": "pro",
"sound": "off"
}
```
### Video reference and continuation
```json theme={null}
{
"model_name": "kling-o1",
"prompt": "Use the action and camera movement from <<>> to generate the immediately following shot",
"video_list": [
{
"video_url": "https://example.com/reference.mp4",
"refer_type": "feature",
"keep_original_sound": "no"
}
],
"mode": "pro",
"aspect_ratio": "16:9",
"duration": "5",
"sound": "off"
}
```
### First and end frames
```json theme={null}
{
"model_name": "kling-o1",
"prompt": "Transition naturally from morning to night while preserving the building structure",
"image_list": [
{"image_url": "https://example.com/first.png", "type": "first_frame"},
{"image_url": "https://example.com/end.png", "type": "end_frame"}
],
"mode": "pro",
"duration": "6",
"sound": "off"
}
```
## FAQ
### What input media is supported?
* Up to 7 JPG/JPEG/PNG images, at least 300 px per side and no more than 10 MB each.
* At most one MP4/MOV video, at least 3 seconds, up to 2K and 200 MB.
* A multi-image element can be built from up to 4 different views.
* With a reference video, images and elements total at most 4; without video, they total at most 7.
### How do prompts reference media?
Use `<<>>`, `<<>>`, and `<<>>` in array order. The numbering must match the corresponding list.
### What restrictions apply to frames and video editing?
End-frame-only generation is not supported. An end frame requires a first frame, and first/end-frame mode cannot use elements. Source-video editing with `refer_type: "base"` also cannot use frames, and its output duration follows the source.
### Which resolutions and durations are supported?
`std` is 720p and `pro` is 1080p; 4K is not supported. Output duration is 3–10 seconds except for source-video editing.
## Query Task (Single)
`GET /kling/v1/videos/omni-video/{task_id}`
```bash cURL theme={null}
curl --request GET \
--url https://www.anyfast.ai/kling/v1/videos/omni-video/860260753860210752 \
--header 'Authorization: Bearer YOUR_API_KEY'
```
Task ID returned by the create endpoint.
### Response fields
Business status code; `0` indicates success.
Response message.
Request trace ID.
Task ID.
`submitted`, `processing`, `succeed`, or `failed`.
Failure reason or status details.
Generated videos.
Video ID.
Video URL without a watermark.
Watermarked video URL.
Actual video duration.
Whether the platform watermark is enabled.
Final resource-unit deduction.
Final balance quota deduction.
Creation timestamp in milliseconds.
Update timestamp in milliseconds.
```json 200 theme={null}
{
"code": 0,
"message": "SUCCEED",
"request_id": "req_123456789",
"data": {
"task_id": "860260753860210752",
"task_status": "succeed",
"task_status_msg": "",
"task_result": {
"videos": [
{
"id": "video_123456789",
"url": "https://example.com/generated.mp4",
"watermark_url": "https://example.com/generated-watermark.mp4",
"duration": "7"
}
]
},
"watermark_info": {"enabled": false},
"final_unit_deduction": "1",
"final_balance_deduction": {"quota": "0"},
"created_at": 1773130665000,
"updated_at": 1773130765000
}
}
```
Kling removes image and video URLs from task results after 30 days. Download and save generated media before the URLs expire.
## Element APIs
| Operation | Method and path |
| --------------------- | ---------------------------------------- |
| Create custom element | `POST /kling/v1/general/custom-elements` |
| List custom elements | `GET /kling/v1/general/custom-elements` |
| List preset elements | `GET /kling/v1/general/presets-elements` |
| Delete custom element | `POST /kling/v1/general/delete-elements` |
This section strictly documents the Kling Legacy Element API.
## Create Element
`POST /kling/v1/general/custom-elements`
```bash cURL theme={null}
curl --request POST \
--url https://www.anyfast.ai/kling/v1/general/custom-elements \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"element_name": "series-character",
"element_description": "The lead character",
"element_frontal_image": "https://example.com/front.png",
"element_refer_list": [
{"image_url": "https://example.com/side.png"},
{"image_url": "https://example.com/close-up.png"}
],
"tag_list": [{"tag_id": "o_102"}]
}'
```
```json 200 theme={null}
{
"code": 0,
"message": "SUCCEED",
"request_id": "request-id",
"data": {
"element_id": 863121023120580662,
"element_name": "series-character",
"element_description": "The lead character",
"element_frontal_image": "https://example.com/front.png",
"element_refer_list": [
{"image_url": "https://example.com/side.png"}
],
"tag_list": [
{"id": "o_102", "name": "Character", "description": "string"}
],
"owned_by": "creator-id"
}
}
```
### Request headers
Bearer authentication in the form `Bearer YOUR_API_KEY`.
Use `application/json`.
### Request body
Element name, up to 20 characters.
Element description, up to 100 characters.
Front-view image URL or raw Base64 without a data URI prefix.
One to three alternate-angle or close-up images.
Reference image URL or raw Base64.
Element tags.
`o_101` trend, `o_102` character, `o_103` animal, `o_104` item, `o_105` costume, `o_106` scene, `o_107` effect, or `o_108` other.
Images support JPG, JPEG, and PNG, up to 10 MB each. Both dimensions must be at least 300 px and the aspect ratio must be from 1:2.5 to 2.5:1.
### Create response
`0` indicates success.
Response message.
Request trace ID.
Element ID used in O1 `element_list`.
Element name.
Element description.
Front-view image.
Alternate-angle reference images.
Element tags.
Element owner; `kling` identifies an official preset.
## Query Custom Element (List)
`GET /kling/v1/general/custom-elements?pageNum=1&pageSize=30`
```bash cURL theme={null}
curl --request GET \
--url 'https://www.anyfast.ai/kling/v1/general/custom-elements?pageNum=1&pageSize=30' \
--header 'Authorization: Bearer YOUR_API_KEY'
```
Page number from 1 to 1,000.
Items per page from 1 to 500.
The response contains `code`, `message`, `request_id`, and `data[]`. Each item includes the create-response element fields and `final_unit_deduction`.
## Query Preset Element (List)
`GET /kling/v1/general/presets-elements?pageNum=1&pageSize=30`
```bash cURL theme={null}
curl --request GET \
--url 'https://www.anyfast.ai/kling/v1/general/presets-elements?pageNum=1&pageSize=30' \
--header 'Authorization: Bearer YOUR_API_KEY'
```
Page number from 1 to 1,000.
Items per page from 1 to 500.
The response matches the custom-element list without `final_unit_deduction`. Official presets use `owned_by: "kling"`.
## Delete Custom Element
`POST /kling/v1/general/delete-elements`
```bash cURL theme={null}
curl --request POST \
--url https://www.anyfast.ai/kling/v1/general/delete-elements \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{"element_id": "863121023120580662"}'
```
Custom element ID to delete. Official preset elements cannot be deleted.
```json 200 theme={null}
{
"code": 0,
"message": "SUCCEED",
"request_id": "request-id"
}
```