Video Crafter

video-crafter

Video Crafter is open diffusion model for high-quality video generation

A100 80GB

Fast Inference

REST API

Try in Console API Docs Examples

Model Information

Response Time~192 sec

StatusActive

Version

0.0.1

Updated29 days ago

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": "video-crafter",
            "version": "0.0.1",
            "input": {
  "fps": "28",
  "seed": null,
  "steps": "50",
  "width": "1024",
  "height": "576",
  "prompt": "your prompt here"
}
        }
    )
    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: ~192 seconds
Rate limit: 60 requests/minute
Concurrent requests: 10 maximum
Use long-polling to check prediction status until completion

Overview

Video Crafter is a state-of-the-art generative model designed to create high-quality videos based on text prompts and customizable parameters. With its intuitive structure and advanced algorithms, it enables users to produce tailored video outputs efficiently. Video Crafter supports various input configurations, providing users with flexibility and control over video generation.

Technical Specifications

Generative Model: Advanced video synthesis using cutting-edge AI techniques.
Customization: Allows users to fine-tune width, height, and frame rates for tailored outputs.
Reproducibility: Incorporates seed functionality to replicate results effectively.
Output Quality: Governed by steps, where higher values yield better detail at the cost of longer generation times.

Key Considerations

Input Compatibility:

Ensure all parameters align with your system's capabilities to avoid errors or suboptimal results.

Performance Trade-offs:

Higher resolution and step count can significantly increase generation time and resource consumption.

Reproducibility:

Minor changes in prompts or parameters can lead to drastically different results. Document settings for consistency.

Output Length:

While the model doesn’t limit duration explicitly, parameter configurations like fps and steps directly impact perceived length and smoothness.

Tips & Tricks

Prompt for Video Crafter:

Create descriptive and vivid prompts. For example:
- Poor: “A city.”
- Better: “A futuristic cityscape at sunset with flying cars.”
Avoid redundancy or overly long descriptions that dilute focus.

Width & Height:

Recommended aspect ratios:
- Landscape: Use a higher width than height (e.g., 1920x1080).
- Portrait: Height > Width (e.g., 1080x1920).
- Square: Equal width and height (e.g., 1080x1080).
For optimal performance, ensure width and height are divisible by 64 to align with model requirements.

Steps for Video Crafter:

Low steps (10-25): Fast generation, lower quality.
Medium steps (30-50): Balanced quality and speed.
High steps (50+): Superior quality, but longer rendering times.

FPS:

15 fps: Suitable for slower, artistic visuals.
24 fps: Standard for cinematic quality.
30 fps+: Ideal for smooth, high-quality videos.

Seed:

Use a fixed seed to replicate successful results.
For creative exploration, randomize seeds to discover new variations.

Capabilities

Dynamic Content Creation: Generate videos tailored to specific prompts and visual styles with Video Crafter.

Customizable Output: Adjust resolution, smoothness, and steps to match artistic or functional needs.

Reproducibility: Ensure consistent outputs for similar input configurations.

Creative Exploration: Experiment with diverse styles and variations using adjustable parameters.

What can I use for?

Media Production:

Quickly generate storyboards or conceptual animations for films or ads.

Artistic Projects:

Create unique video art based on detailed textual descriptions.

Educational Content:

Produce illustrative videos for training or instructional materials.

Marketing Campaigns:

Design engaging and personalized video advertisements.

Things to be aware of

Generate a futuristic landscape with dynamic lighting by adjusting the steps and seed values with Video Crafter.

Experiment with artistic frame rates (e.g., 15 fps) to create surreal animations.

Combine multiple prompts to produce a storyboard-like sequence.

Explore the impact of aspect ratio by switching between landscape, portrait, and square dimensions.

Limitations

Resource-Intensive:

Generating high-resolution videos with high step counts requires significant computational power.

Output Diversity:

While robust, outputs may occasionally deviate from the intended prompt or exhibit artifacts, especially with complex or abstract descriptions.

Output Format: MP4

Related AI Models

Wan 2.1-1.3B

wan-2-1-1-3b

Text to Video

Hailuo Video

hailuo-video

Text to Video

Luma Ray2 540p

ray-2-540p

Text to Video

Gen-2 by Runway

runway

Text to Video