Generate Stunning Images with the hadasadler/newman Cognitive Actions

24 Apr 2025
Generate Stunning Images with the hadasadler/newman Cognitive Actions

In the realm of artificial intelligence, image generation has become a fascinating frontier. The hadasadler/newman spec provides developers with powerful Cognitive Actions to create images using advanced techniques such as inpainting and image-to-image transformations. By leveraging these pre-built actions, you can harness the power of AI to generate high-quality images tailored to your specific requirements, whether for creative projects, marketing materials, or personal use.

Prerequisites

Before diving into the Cognitive Actions, ensure you have the following:

  • An API key for the Cognitive Actions platform. This key is crucial for authenticating your requests.
  • Basic understanding of how to make API calls, particularly using JSON format for input and output.

Authentication typically involves passing the API key in the headers of your HTTP requests.

Cognitive Actions Overview

Generate Image with Inpainting or Image-to-Image

This action allows you to generate images using inpainting or image-to-image techniques, offering a variety of customizable parameters such as aspect ratio, dimensions, model selection, output quality, and more. It supports both fast generation modes for quicker outputs and high-quality settings for detailed images.

Input

The input for this action is structured around the CompositeRequest schema, which requires the following fields:

  • prompt (required): A description of the image you want to generate.
  • mask: A URI for an image mask used in inpainting mode.
  • seed: An integer for random number generation.
  • image: URI of an input image for image-to-image or inpainting modes.
  • width: Width of the generated image in pixels (256 to 1440).
  • height: Height of the generated image in pixels (256 to 1440).
  • fastMode: Enable for faster processing.
  • imageFormat: Output image format (webp, jpg, png).
  • modelChoice: Select the model for inference (dev, schnell).
  • outputCount: Number of images to generate (1 to 4).
  • imageQuality: Quality level for output images (0 to 100).
  • guidanceScale: Scale for the diffusion process (0 to 10).
  • inferenceSteps: Number of denoising steps (1 to 50).
  • promptStrength: Strength of the prompt when using img2img (0 to 1).
  • imageAspectRatio: Aspect ratio for the generated image.
  • loraWeightsScale: Adjusts strength of the main LoRA application.
  • outputResolution: Approximate number of megapixels.
  • additionalLoraScale: Strength of the additional LoRA application.
  • additionalLoraWeights: Load additional LoRA weights.
  • safetyCheckerDisabled: Bypass the safety checker for images.

Example Input:

{
  "prompt": "newman  A six-month-old chubby baby sitting on the floor at home surrounded by a pile of envelopes and toys, with a mischievous smirk, holding a puffy envelope as if he’s about to pull off a little prank.",
  "fastMode": false,
  "imageFormat": "jpg",
  "modelChoice": "dev",
  "outputCount": 1,
  "imageQuality": 80,
  "guidanceScale": 3,
  "inferenceSteps": 28,
  "promptStrength": 0.8,
  "imageAspectRatio": "16:9",
  "loraWeightsScale": 1,
  "outputResolution": "1",
  "additionalLoraScale": 1
}

Output

The action typically returns a list of image URLs generated based on the input parameters.

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/e68f370e-52ab-4a23-b98a-46361d7a400d/5b5c4652-0b05-42cc-90e9-41cb37c8fc1e.jpg"
]

Conceptual Usage Example (Python)

Below is a conceptual Python code snippet demonstrating how to invoke this action using a hypothetical Cognitive Actions execution endpoint:

import requests
import json

# Replace with your Cognitive Actions API key and endpoint
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute" # Hypothetical endpoint

action_id = "e8f79b36-74f8-4582-8583-069f529b72f2" # Action ID for Generate Image with Inpainting or Image-to-Image

# Construct the input payload based on the action's requirements
payload = {
    "prompt": "newman  A six-month-old chubby baby sitting on the floor at home surrounded by a pile of envelopes and toys, with a mischievous smirk, holding a puffy envelope as if he’s about to pull off a little prank.",
    "fastMode": false,
    "imageFormat": "jpg",
    "modelChoice": "dev",
    "outputCount": 1,
    "imageQuality": 80,
    "guidanceScale": 3,
    "inferenceSteps": 28,
    "promptStrength": 0.8,
    "imageAspectRatio": "16:9",
    "loraWeightsScale": 1,
    "outputResolution": "1",
    "additionalLoraScale": 1
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json"
}

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json={"action_id": action_id, "inputs": payload} # Hypothetical structure
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully:")
    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: {e.response.text}")

In this code snippet, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id corresponds to the "Generate Image with Inpainting or Image-to-Image" action. The payload is structured according to the input schema, ensuring you provide all required fields.

Conclusion

By utilizing the hadasadler/newman Cognitive Actions, you can effortlessly generate high-quality images tailored to your specifications. With options for customization, including aspect ratios, image quality, and different models, the possibilities are nearly endless. Consider experimenting with these actions to create unique visuals for your projects, and explore additional use cases to take full advantage of this powerful toolset. Happy coding!

← Back to All Articles