Kling AI

Kling 2.5 Turbo

Kling 2.5 Turbo separates generation by input type and processing mode. Text-to-video builds the entire shot from the prompt, while image-to-video treats the uploaded image as the starting visual state and uses the prompt to direct motion and camera behavior.

How to use this model

  1. 1

    Choose the T2V variant for a text-only concept or an I2V variant when an exact opening composition matters.

  2. 2

    In the prompt, describe subject motion separately from camera motion so the two directions do not conflict.

  3. 3

    When animating an image, use a clean source with enough space in the direction where the subject or camera should move.

  4. 4

    Select Standard or Pro according to the available version, then set duration and aspect ratio for the final placement.

2.5 Turbo ProAvailable

Kling 2.5 Turbo

Choose the quality tier directly. Omnipix sends text-only requests to Pro text-to-video and automatically switches to image-to-video when a start image is supplied; Standard is available for image-to-video only.

Try now

Properties

  • Text or image input determines the generation workflow automatically.
  • Standard is the lower-cost image-to-video tier; Pro supports both workflows.

How to use this model

  1. 1

    Choose Pro or Standard quality.

  2. 2

    Write the shot prompt and optionally upload a start image.

  3. 3

    For Pro, optionally add an end image to guide the final frame.

  4. 4

    Select duration and review motion continuity.

Examples

A reviewed video output created with 2.5 Turbo Pro - Image to Video in Omnipix.
A reviewed video output created with 2.5 Turbo Pro - Text to Video in Omnipix.
A reviewed video output created with 2.5 Turbo Standard - Image to Video in Omnipix.

Parameters

Prompt

prompt
Required

Text prompt for generation. Max length: 2500 characters

Duration

duration
Optional

The duration of the generated video in seconds Default value: "5"

Default: 5

  • 5

  • 10

Negative Prompt

negative_prompt
Optional

What you don't want in the generated video. default value: blur, distort, and low quality

Default: blur, distort, and low quality

CFG Scale

cfg_scale
Optional

The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt. Default value: 0.5

Default: 0.5Range: 0 – 1

Quality

mode
Optional

Pro supports text-to-video, image-to-video, and end-frame interpolation. Standard is the lower-cost image-to-video option.

Default: pro

  • Pro

  • Standard

Aspect Ratio

aspect_ratio
Optional

Aspect ratio for text-to-video generation. Image-to-video uses the supplied start image.

  • 16:9

  • 1:1

  • 9:16

Start Image

image_url
Optional

URL of the image to be used for the start frame

End Image

tail_image_url
Optional

URL of the image to be used for the end frame