kling/kling-video-v3-pro
Kling V3 Pro turns a still image into cinematic video with strong motion consistency, and can generate native audio in the same pass.
Model Inputs
Text prompt describing how the video unfolds from the first frame
First frame of the video (.jpg/.png/.webp)
Drag & drop a file here, or click to select
Supported formats: JPG, PNG, WEBP (max 10MB)
The duration of the generated video in seconds
Whether to generate native audio for the video. Supports Chinese and English voice output. Other languages are automatically translated to English. For English speech, use lowercase letters; for acronyms or proper nouns, use uppercase.
Optional last frame (.jpg/.png/.webp). When set, the clip is generated as a transition from the start image to this one
Drag & drop a file here, or click to select
Supported formats: JPG, PNG, WEBP (max 10MB)
The type of multi-shot video generation. 'intelligent' lets the model automatically determine shot structure.
Things to keep out of the video
The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt.
Result
Pricing
kling/
kling-video-v3-pro
You are billed for what this model actually produces — by the second — so a shorter request costs less and you never pay for idle time.
For example, a request of 60 seconds costs about $10.08.
You are charged for what the request actually produces, so the final amount scales with the seconds you use.
| Provider | Price ($) | Saving (%) |
|---|---|---|
| Synexa | $0.112 | - |
| fal | $0.84 | 86.7% |
Comparison prices are for an equivalent request.
Readme
Kling V3 Pro is the latest generation of the Kling image-to-video family. It animates a start image into a clip of 3-15 seconds with improved visual fidelity and motion consistency over V2.
Native audio generation is on by default and supports Chinese and English voice output; other languages are translated to English automatically. Supply an end image to generate a transition between two frames, and use the negative prompt to suppress artefacts.
Multi-shot storyboards and character/element references exist upstream but are not exposed here, because they take nested object payloads.