Kling v1 Standard Image to Video

kling-v1-standard-image-to-video

Kling v1 Standard Image to Video converts images into smooth, high-quality videos.

Fast Inference

REST API

Try in Console API Docs Examples

Model Information

Response Time~270 sec

StatusActive

Version

0.0.1

Updatedabout 14 hours ago

Live Demo

Average runtime: ~270 seconds

Input

Configure model parameters

Prompt

The prompt for the video

A thrilling first-person view of soaring atop a mighty dragon, weaving through towering stone spires. Its massive wings beat powerfully, slicing through the air and sending gusts spiraling outward. The leathery membranes ripple before snapping downward, propelling the creature forward. Wisps of mist swirl as the dragon banks sharply, its sinuous tail whipping past.The rider grips the dragon’s reins or rough scales, feeling its raw power. Sunlight filters through the clouds, casting a golden glow on its shimmering crimson, emerald, or iridescent blue scales. Embers flicker from its nostrils, or a faint magical aura pulses along its spine.As the dragon dives, its wings fold slightly before unfurling in a controlled descent, the wind roaring in the rider’s ears. Towers rush past, their ancient stonework cracked and moss-covered, before the dragon pulls up gracefully, wings snapping wide to catch the air. The landscape below unfolds like a painting—rolling hills, dense forests, and distant peaks drifting in the clouds. The entire scene brims with motion, light, and shadow, making the flight exhilarating and immersive.

Image URL

URL of the image to be used for the video

File upload is currently disabled

Output

View generated results

Result

Preview, share or download your results with a single click.

Each execution costs $0.14 With $1 you can run this model about 7 times.

API Reference

View Full Documentation

Prerequisites

Create an API Key from the Eachlabs Console
Install the required dependencies for your chosen language (e.g., requests for Python)

API Integration Steps

1. Create a Prediction

Send a POST request to create a new prediction. This will return a prediction ID that you'll use to check the result. The request should include your model inputs and API key.

import requests
import time

API_KEY = "YOUR_API_KEY"  # Replace with your API key
HEADERS = {
    "X-API-Key": API_KEY,
    "Content-Type": "application/json"
}

def create_prediction():
    response = requests.post(
        "https://api.eachlabs.ai/v1/prediction/",
        headers=HEADERS,
        json={
            "model": "kling-v1-standard-image-to-video",
            "version": "0.0.1",
            "input": {
  "static_mask_url": "your static image url here",
  "tail_image_url": "your tail image url here",
  "cfg_scale": 0.5,
  "negative_prompt": "blur, distort, and low quality",
  "aspect_ratio": "16:9",
  "duration": 5,
  "image_url": "your image url here",
  "prompt": "your prompt here"
},
            "webhook_url": ""
        }
    )
    prediction = response.json()
    
    if prediction["status"] != "success":
        raise Exception(f"Prediction failed: {prediction}")
    
    return prediction["predictionID"]

2. Get Prediction Result

Poll the prediction endpoint with the prediction ID until the result is ready. The API uses long-polling, so you'll need to repeatedly check until you receive a success status.

def get_prediction(prediction_id):
    while True:
        result = requests.get(
            f"https://api.eachlabs.ai/v1/prediction/{prediction_id}",
            headers=HEADERS
        ).json()
        
        if result["status"] == "success":
            return result
        elif result["status"] == "error":
            raise Exception(f"Prediction failed: {result}")
        
        time.sleep(1)  # Wait before polling again

3. Complete Example

Here's a complete example that puts it all together, including error handling and result processing. This shows how to create a prediction and wait for the result in a production environment.

try:
    # Create prediction
    prediction_id = create_prediction()
    print(f"Prediction created: {prediction_id}")
    
    # Get result
    result = get_prediction(prediction_id)
    print(f"Output URL: {result['output']}")
    print(f"Processing time: {result['metrics']['predict_time']}s")
except Exception as e:
    print(f"Error: {e}")

Additional Information

The API uses a two-step process: create prediction and poll for results
Response time: ~270 seconds
Rate limit: 60 requests/minute
Concurrent requests: 10 maximum
Use long-polling to check prediction status until completion

Overview

Kling v1 Standard Image to Video generates short video sequences by transforming a static input image into dynamic motion guided by a descriptive text prompt. Kling v1 Standard Image to Video allows for customizable generation through various visual parameters including duration, aspect ratio, and auxiliary image inputs. Kling v1 Standard Image to Video is designed to create natural motion continuity between frames while preserving the original content structure of the image.

Technical Specifications

Kling v1 Standard Image to Video leverages advanced diffusion-based temporal modeling to generate consistent frame-to-frame motion.

Motion vectors are inferred from both prompt semantics and source image layout.

Designed to minimize flicker and artifacts by balancing global scene context with local pixel stability.

Supports frame-level interpolation and motion estimation between image pairs when tail_image_url is used.

Dynamic masking is internally applied to stabilize high-frequency regions unless overridden via static_mask_url.

Ensure the input image has a clear subject with minimal noise to maintain focus in motion rendering.

When using tail_image_url, select images with similar lighting and subject perspective to the main image_url for smoother transitions.

Keep prompts simple and descriptive; overly complex prompts can result in disjointed visuals.

Using a static mask (static_mask_url) can help maintain background or subject stability, depending on the use case.

Videos are currently limited to 5 or 10 seconds; longer durations are not supported.

Aspect ratio should match the subject orientation to avoid distortion.

Key Considerations

Input image quality directly affects the output. Low-resolution or overly compressed images may produce blurry or jittery results.

Prompts should be focused on motion, mood, or transformation. Avoid cluttering the prompt with scene descriptions already present in the image.

If both tail_image_url and static_mask_url are provided, the model prioritizes motion blending and overrides internal motion smoothing logic.

Videos are not audio-synced and contain no sound.

Legal Information for Kling v1 Standard Image to Video

By using this Kling v1 Standard Image to Video, you agree to:

Kling Privacy
Kling SERVICE AGREEMENT

Tips & Tricks

prompt
Use clear, action-oriented phrases (e.g., “a woman turning around slowly”, “clouds drifting across the sky”). Avoid abstract or poetic language.

cfg_scale
Controls adherence to the prompt.

Recommended value: 0.7
Lower values (0.3–0.5): more freedom, creative outputs.
Higher values (0.8–1): stricter adherence to prompt, but risk of less natural motion.

duration

Options: 5 or 10 seconds
Shorter durations result in more focused, stable animations.
For complex prompts, use 10 seconds to allow the model more frames to interpret motion.

aspect_ratio

Options: 16:9, 9:16, 1:1
Match subject framing:
- 16:9 for landscape
- 9:16 for portraits
- 1:1 for centered subjects

image_url
Use high-quality images with the subject in the center. Plain or soft backgrounds produce cleaner animation.

tail_image_url
Adds dynamic ending. Use when transitioning between scenes or actions. Should visually align with the main image.

static_mask_url
Use this if part of the image should remain static. Ideal for keeping the background unchanged while animating the foreground.

negative_prompt
Use to exclude unwanted elements (e.g., “blurry, distorted, extra limbs”).

Capabilities

Transforms static images into short animated sequences.

Allows dynamic motion customization via textual descriptions.

Supports motion continuity between two input images.

Enables foreground/background isolation through masking.

Generates content with consistent subject focus and lighting retention.

What can I use for?

Creating animated portraits or character loops.

Enhancing still images with realistic motion for digital content.

Producing short, looping visual stories for creative visuals.

Visualizing mood or atmosphere changes (e.g., lighting shifts, subtle motion).

Crafting seamless visual transitions between two related scenes.

Things to be aware of

Animate a photograph of a person with a prompt like:
"a person smiling and tilting their head"

Combine two images (main and tail) with:

image_url: A person standing still
tail_image_url: Same person starting to walk
Prompt: "the person begins to walk forward"

Use static_mask_url to keep a building steady while animating the sky:

Prompt: "clouds slowly moving"
static_mask_url: mask over the building

Limitations

Limited to 5 or 10 seconds of output.

Model may struggle with complex or overlapping motion instructions.

Background artifacts may appear when subject edges are unclear.

Does not support facial lip-sync or precise expression control.

No support for audio integration.

Output Format: MP4

Related AI Models