Create Stunning Images Effortlessly with ElasticDiffusion Cognitive Actions

24 Apr 2025
Create Stunning Images Effortlessly with ElasticDiffusion Cognitive Actions

In today's digital landscape, generating high-quality images programmatically opens up a multitude of possibilities for developers. The ElasticDiffusion API provides a powerful toolset through Cognitive Actions that allows you to create stunning images without the need for retraining models. Leveraging pretrained text-to-image diffusion models, these actions ensure superior image coherence and offer flexibility across various use cases. In this article, we will explore how to integrate the Generate Arbitrary Size Images action into your applications.

Prerequisites

Before diving into the integration, ensure you have the following:

  • An API key for accessing the ElasticDiffusion Cognitive Actions.
  • Basic knowledge of making HTTP requests and handling JSON data in your programming environment.
  • Familiarity with Python to follow along with the conceptual examples provided.

Authentication typically involves passing your API key as a Bearer token in the request headers.

Cognitive Actions Overview

Generate Arbitrary Size Images

The Generate Arbitrary Size Images action allows developers to create high-quality images of varying sizes and aspect ratios by providing a textual prompt. This action is part of the image-generation category, making it an essential tool for applications focused on generating visual content.

Input

The input schema for this action consists of several properties, with the prompt being mandatory. Here’s a breakdown:

  • prompt (required): A detailed textual description to guide the image generation process.
  • seed (optional): An integer for initializing the random number generator to ensure reproducibility (default is 0).
  • batchSize (optional): The number of images to generate in one batch (default is 16).
  • imageWidth (optional): The width of the generated image in pixels; must be between 128 and 2048 (default is 768).
  • imageHeight (optional): The height of the generated image in pixels; must be between 128 and 2048 (default is 768).
  • cosineScale (optional): A scaling factor for cosine function during processing (default is 10).
  • guidanceScale (optional): A scaling factor influencing adherence to the prompt (default is 10).
  • negativePrompt (optional): A string of attributes to avoid in the generated image (default is an empty string).
  • resamplingSteps (optional): Number of resampling steps during generation (default is 10).
  • numberOfInferenceSteps (optional): Steps the model will take during inferencing; must be between 30 and 100 (default is 50).
  • resamplingNewPercentage (optional): Percentage of pixels to be resampled as new during generation (default is 0.3).
  • reducedResolutionGuidanceScale (optional): A scaling factor used at reduced resolution levels (default is 400).

Example Input:

{
  "seed": 0,
  "prompt": "Envision a portrait of a cute cat, her face is framed by a blue headscarf with muted tones of rust and cream. Her eyes are blue like faded denim. Her attire, simple yet dignified",
  "batchSize": 16,
  "imageWidth": 1080,
  "cosineScale": 10,
  "imageHeight": 1920,
  "guidanceScale": 10,
  "negativePrompt": "blurry, ugly, poorly drawn, deformed",
  "resamplingSteps": 7,
  "numberOfInferenceSteps": 50,
  "resamplingNewPercentage": 0.3,
  "reducedResolutionGuidanceScale": 1000
}

Output

Upon successful execution, the action returns a URL pointing to the generated image. Here’s an example of a typical output:

Example Output:

https://assets.cognitiveactions.com/invocations/289bb89e-e9bf-402d-84f1-ef5904d45b37/8fc78a5d-3495-4ae1-a122-343ae68eb1a0.png

Conceptual Usage Example (Python)

Here’s how a developer might call the Generate Arbitrary Size Images action using Python:

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 = "80a5a68b-c7af-4b71-85c2-d2dc2faed840" # Action ID for Generate Arbitrary Size Images

# Construct the input payload based on the action's requirements
payload = {
    "seed": 0,
    "prompt": "Envision a portrait of a cute cat, her face is framed by a blue headscarf with muted tones of rust and cream. Her eyes are blue like faded denim. Her attire, simple yet dignified",
    "batchSize": 16,
    "imageWidth": 1080,
    "cosineScale": 10,
    "imageHeight": 1920,
    "guidanceScale": 10,
    "negativePrompt": "blurry, ugly, poorly drawn, deformed",
    "resamplingSteps": 7,
    "numberOfInferenceSteps": 50,
    "resamplingNewPercentage": 0.3,
    "reducedResolutionGuidanceScale": 1000
}

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, the action ID and input payload are clearly illustrated. The endpoint URL and request structure are hypothetical, serving as a guide for developers to structure their own requests.

Conclusion

The ElasticDiffusion Cognitive Actions provide a straightforward and powerful way to generate stunning images from textual descriptions. By utilizing the Generate Arbitrary Size Images action, developers can create unique visual content tailored to their applications. As you explore this functionality, consider integrating it into various use cases, from creative projects to dynamic content generation. Happy coding!

← Back to All Articles