Skip to main content
POST
/
v3
/
async
/
kling-2.5-turbo-i2v
Kling V2.5 Turbo Image to Video
curl --request POST \
  --url https://api.novita.ai/v3/async/kling-2.5-turbo-i2v \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "image": "<string>",
  "prompt": "<string>",
  "duration": "<string>",
  "cfg_scale": 123,
  "mode": "<string>",
  "negative_prompt": "<string>"
}
'
{
  "task_id": "<string>"
}

Documentation Index

Fetch the complete documentation index at: https://novita.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

Kling 2.5 Turbo Image-to-Video can generate cinematic, high-quality videos from a single image and text prompt. It features a new text-temporal engine, enhanced motion rendering, and faster inference speed, resulting in smooth action and content that closely matches the given prompt.
This is an asynchronous API; only the task_id will be returned. You should use the task_id to request the Task Result API to retrieve the video generation results.

​
Request Headers

​
Content-Type
string
required
Supports: application/json
​
Authorization
string
required
Bearer authentication format, for example: Bearer {{API Key}}.

​
Request Body

​
image
string
required
The initial frame image for the video. Supported formats: .jpg, .jpeg, .png. Image size must not exceed 10MB. Minimum resolution: 300x300 pixels.
​
prompt
string
required
Positive prompt text for video generation. Maximum length: 2500 characters.
​
duration
string
default:"5"
Duration of the generated video (in seconds).Allowed values: 5, 10
​
cfg_scale
number
Controls generation flexibility. Higher values increase adherence to the prompt but reduce creative freedom.Value range: 0 to 1
​
mode
string
default:"pro"
Video generation mode. Allowed value:
  • pro: professional mode
​
negative_prompt
string
Negative prompt to avoid undesired content in the video. Maximum length: 2500 characters.

​
Response

​
task_id
string
required
Use the task_id to request the Task Result API to retrieve the generated outputs.