Kwaivgi Kling V2.5 Turbo Pro Image To Video

Playground

Kling 2.5 Turbo Pro: A cutting-edge solution for image-to-video generation, delivering unmatched motion fluidity, cinematic-grade visuals, and remarkable prompt precision. With its enhanced dynamic effect processing and upgraded text-to-motion responsiveness, it ensures seamless transitions while maintaining exceptional aesthetic fidelity across all styles.

Features

Enhanced dynamic effect processing and upgraded text-to-motion responsiveness:

Text prompt interpretation and timing control have been further improved.

New one-generation model has achieved significant advancements in text interpretation. It has optimized the text structure and internal processing architecture to precisely analyze text instructions that involve multiple steps and causal relationships. This enhances the ability of the model to fully interpret and convert static images into a narrative film with a seamless sequence, ensuring precise control over creative expression.

High-speed dynamic quality and stability:

To generate more dynamic and wide-ranging motion, the new generation model has introduced advanced training methods and strategically adjusted the training data distribution. This enables the model to effectively learn and mimic the physical dynamics of the real world. It can generate actions that were previously difficult to achieve, such as high-speed movements and camera angles. The entire process not only enhances the intensity and range of motion but also ensures smooth and stable video frames, avoiding common glitches and discontinuities that commonly occur in complex dynamics.

Generation inference acceleration:

The new generation model has optimized various generation frameworks to significantly improve overall inference speed. This strategy has reduced generation latency, allowing users to receive high-quality results faster, while balancing efficiency and experience.

Consistency in various styles:

To ensure seamless integration with the original art style, a sophisticated image condition injection technique based on high-density training data is employed. This ensures precise capturing of the visual features of the original work, including color palette, lighting distribution, brush strokes, and overall atmosphere. As a result, even in complex dynamics, each frame of the final product retains a high level of visual consistency with the original image.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result


# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/kwaivgi/kling-v2.5-turbo-pro/image-to-video" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "guidance_scale": 0.5,
    "duration": "5"
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"

Parameters

Task Submission Parameters

Request Parameters

Parameter	Type	Required	Default	Range	Description
image	string	Yes		-	First frame of the video; Supported image formats include.jpg/.jpeg/.png; The image file size cannot exceed 10MB, and the image resolution should not be less than 300*300px, and the aspect ratio of the image should be between 1:2.5 ~ 2.5:1.
prompt	string	Yes		-	Text prompt for generation; Positive text prompt; Cannot exceed 2500 characters.
negative_prompt	string	No		-	Negative text prompt; Cannot exceed 2500 characters.
guidance_scale	number	No	0.5	0.00 ~ 1.00	The guidance scale to use for the generation.
duration	string	No	5	5, 10	The duration of the generated media in seconds.

Response Parameters

Parameter	Type	Description
code	integer	HTTP status code (e.g., 200 for success)
message	string	Status message (e.g., “success”)
data.id	string	Unique identifier for the prediction, Task Id
data.model	string	Model ID used for the prediction
data.outputs	array	Array of URLs to the generated content (empty when status is not `completed`)
data.urls	object	Object containing related API endpoints
data.urls.get	string	URL to retrieve the prediction result
data.has_nsfw_contents	array	Array of boolean values indicating NSFW detection for each output
data.status	string	Status of the task: `created`, `processing`, `completed`, or `failed`
data.created_at	string	ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”)
data.error	string	Error message (empty if no error occurred)
data.timings	object	Object containing timing details
data.timings.inference	integer	Inference time in milliseconds

Result Request Parameters

Kwaivgi Kling V2.1 T2V Master Kwaivgi Kling V2.5 Turbo Pro Text To Video