video-generation

Create Stunning Videos Effortlessly with Mochi 1

27 Apr 2025
Create Stunning Videos Effortlessly with Mochi 1

In the rapidly evolving world of digital media, the ability to generate high-quality video content quickly and efficiently is invaluable. Enter Mochi 1, a state-of-the-art video generation model that harnesses advanced techniques to produce high-fidelity videos with remarkable adherence to user-defined prompts. This innovative model utilizes an Asymmetric Diffusion Transformer architecture, allowing developers to create visually stunning videos that meet specific requirements without the need for extensive manual editing.

Whether you are a content creator, marketer, or developer looking to enhance your projects with engaging video content, Mochi 1 simplifies the video production process, enabling you to focus on creativity while leveraging cutting-edge technology. Use cases range from generating promotional videos, educational content, to artistic expressions, making it a versatile tool for various industries.

Prerequisites

To get started with Mochi 1, you'll need a Cognitive Actions API key and a basic understanding of how to make API calls.

Generate High-Fidelity Videos

The "Generate High-Fidelity Videos" action allows users to create high-quality videos that accurately reflect their creative vision. By providing a detailed prompt, developers can guide the model to produce videos that align closely with their expectations.

Input Requirements

To utilize this action effectively, you will need to provide the following inputs:

  • Seed: An integer to initialize the random number generator for reproducible results.
  • Prompt: A descriptive string that details the desired output. For example, "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."
  • Guidance Scale: A numeric value from 0.1 to 10 that influences the model's adherence to the prompt, with a default of 6.
  • Number of Frames: The total number of frames to be generated, ranging from 30 to 170, with a default of 163.
  • Frames Per Second: Specifies the output frame rate, ranging from 10 to 60, with a default of 30.
  • Number of Inference Steps: Defines the number of steps in the inference process, ranging from 10 to 200, with a default of 64.

Expected Output

The expected output is a high-fidelity video that meets the specifications provided in the input. For example, a generated video might be accessible via a link such as: https://assets.cognitiveactions.com/invocations/1d5a4778-3b13-4d5f-ab46-cffd7d5cdbd9/eac26f3b-c4aa-46b4-9fc4-5707781676d8.mp4

Use Cases for this Specific Action

  • Marketing and Advertising: Create captivating promotional videos that highlight products or services.
  • Educational Content: Generate informative and engaging video lessons or explainer videos for online courses.
  • Artistic Projects: Experiment with creative video concepts that push the boundaries of visual storytelling.
  • Social Media Content: Quickly produce high-quality videos tailored for platforms like Instagram, TikTok, or YouTube.
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "81e56ffa-e6a6-43aa-9bb5-84b578381f04" # Action ID for: Generate High-Fidelity Videos

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "prompt": "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k.",
  "guidanceScale": 5.5,
  "numberOfFrames": 121,
  "framesPerSecond": 24,
  "numberOfInferenceSteps": 30
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body (non-JSON): {e.response.text}")
    print("------------------------------------------------")

Conclusion

Mochi 1 revolutionizes the way developers approach video content creation, offering a powerful tool for generating stunning videos with minimal effort. Its versatility opens up numerous possibilities across different fields, from marketing to education and beyond. By leveraging the capabilities of this advanced model, you can enhance your projects, engage your audience, and streamline your workflow.

As you explore the potential of Mochi 1, consider how you can integrate it into your development projects and elevate your video content to new heights.

← Back to All Articles