image-generation

Create Stunning Images Effortlessly with Hart's Image Generation

25 Apr 2025
Create Stunning Images Effortlessly with Hart's Image Generation

In the rapidly evolving world of artificial intelligence, Hart stands out as a powerful tool for developers looking to harness the potential of image generation. With its innovative Hybrid Autoregressive Transformer (HART), Hart allows you to generate high-quality 1024x1024 images efficiently. This advanced model not only combines discrete and continuous tokens to enhance image reconstruction but also outperforms traditional diffusion models, offering faster throughput and lower computational costs.

Imagine the possibilities: from creating unique artwork to generating visual content for apps and websites, Hart's image generation capabilities can streamline your creative process. Whether you're a game developer, a graphic designer, or a content creator, integrating Hart into your workflow can significantly enhance your productivity and creativity.

Prerequisites

To get started, you'll need a Cognitive Actions API key and a basic understanding of making API calls. This will enable you to leverage Hart's powerful image generation features seamlessly.

Generate Image with Hybrid Autoregressive Transformer

Hart's primary action is designed to generate images based on user-defined prompts. This action solves the problem of time-consuming and resource-intensive image creation by providing a fast, efficient, and high-quality output.

Input Requirements

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

  • Seed (optional): An integer for generating pseudo-random numbers. If left blank, the system will generate a random seed.
  • Prompt: A string that describes the image you want to generate (e.g., "An astronaut riding a horse on the moon, oil painting by Van Gogh.").
  • More Smooth: A boolean indicating whether to enable smoother visual transitions to reduce artifacts in the output.
  • Guidance Scale: A number that influences how closely the generated image adheres to the prompt, with a valid range from 1 to 20.
  • Max Token Length: An integer specifying the maximum length of tokens processed from the input prompt, defaulting to 300 tokens.
  • Use Llm System Prompt: A boolean that determines whether to utilize the system-level prompt for language model processing.

Expected Output

Upon successful processing, you will receive a URL linking to the generated image. For example, a request might yield an output such as:

https://assets.cognitiveactions.com/invocations/525724a6-0569-4193-a4a5-74951e1c28fd/f582d67f-ba22-4e8f-a1ef-c78760c601b2.png

Use Cases for this Specific Action

This action is perfect for a myriad of applications, including:

  • Creative Art Generation: Artists and illustrators can use Hart to quickly generate unique pieces based on their creative prompts, serving as inspiration or direct content for their projects.
  • Game Development: Game designers can create assets and backgrounds dynamically, enhancing the visual experience of their games without the extensive resource commitment typically required.
  • Marketing and Advertising: Marketers can generate eye-catching visuals for campaigns and social media posts, allowing for rapid content creation that aligns with their messaging.
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 = "bd68b456-2d20-4620-8cdd-d4e86cf425b1" # Action ID for: Generate Image with Hybrid Autoregressive Transformer

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "prompt": "An astronaut riding a horse on the moon, oil painting by Van Gogh.",
  "moreSmooth": true,
  "guidanceScale": 4.5,
  "maxTokenLength": 300,
  "useLlmSystemPrompt": true
}

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

Hart's image generation capabilities represent a significant leap forward in how developers can create visual content. By leveraging the Hybrid Autoregressive Transformer, you can produce high-quality images efficiently, reducing both time and computational costs. Whether for artistic endeavors, game design, or marketing initiatives, Hart empowers you to bring your visions to life with ease.

To get started, consider experimenting with the different input parameters to see how they affect the generated images. The possibilities are endless, and Hart is here to help you explore them.

← Back to All Articles