Unleashing Artistic Creativity: Integrate Image Generation with SDXL Cognitive Actions

21 Apr 2025
Unleashing Artistic Creativity: Integrate Image Generation with SDXL Cognitive Actions

In the realm of digital creativity, the ferluht/sdxl-pagan API offers a powerful set of Cognitive Actions designed for generating artistic images. One of the standout features of this API is the ability to create images using SDXL, fine-tuned on Vsevolod Ivanov's distinctive painting style. With precise control over image generation and inpainting, developers can achieve stunning results through customizable parameters. This article will guide you through integrating these Cognitive Actions into your applications, making it easy to harness the full potential of image generation.

Prerequisites

Before you begin, ensure you have the following:

  • An API key for the Cognitive Actions platform to authenticate your requests.
  • Basic knowledge of how to structure JSON for API calls.

Authentication typically involves passing your API key in the headers of your HTTP requests, ensuring secure access to the Cognitive Actions services.

Cognitive Actions Overview

Generate Inpainting with SDXL

The Generate Inpainting with SDXL action empowers developers to create artistic images by manipulating existing ones or generating new ones based on textual prompts. This action allows for high customization in image generation, enabling the incorporation of specific artistic styles while controlling various parameters.

  • Category: image-generation

Input

The input schema for this action requires a structured JSON object. Here’s a breakdown of the required and optional fields:

  • mask (string): URI of the input mask for inpaint mode. Black areas will be preserved, and white areas will be inpainted.
  • seed (integer, optional): Random seed for deterministic outputs. If omitted, it generates a random seed.
  • image (string): URI of the input image to be processed.
  • width (integer, default: 1024): Width of the output image in pixels.
  • height (integer, default: 1024): Height of the output image in pixels.
  • prompt (string, default: "An astronaut riding a rainbow unicorn"): A descriptive text prompt guiding the image generation.
  • refine (string, default: "no_refiner"): Style of refinement for image processing (options: "no_refiner", "expert_ensemble_refiner", "base_image_refiner").
  • scheduler (string, default: "K_EULER"): Algorithm for scheduling denoising steps.
  • applyWatermark (boolean, default: true): Whether to apply a watermark to the image.
  • negativePrompt (string): Specifies what should not be included in the image.
  • promptStrength (number, default: 0.8): Influence of the prompt in img2img/inpaint mode, ranging from 0 to 1.
  • numberOfOutputs (integer, default: 1): Number of output images to generate (1-4).
  • refinementSteps (integer, optional): Steps for image refinement when using 'base_image_refiner'.
  • highNoiseFraction (number, default: 0.8): Fraction of noise for refinement using 'expert_ensemble_refiner'.
  • loraAdditiveScale (number, default: 0.6): Scale for LoRA influence.
  • defaultLoraWeights (string): LoRA weights for image generation.
  • guidanceScalingFactor (number, default: 7.5): Guidance scale for adhering to the prompt.
  • safetyCheckerDisabled (boolean, default: false): Indicates if the safety checker is disabled for generated images.
  • numberOfInferenceSteps (integer, default: 50): Denoising steps applied during image generation.

Example Input:

{
  "width": 1024,
  "height": 1024,
  "prompt": "In the style of Vsevolod Ivanov, ancient pagan God standing in the woods and holding a laser sword",
  "refine": "no_refiner",
  "scheduler": "K_EULER",
  "applyWatermark": true,
  "negativePrompt": "",
  "promptStrength": 0.8,
  "numberOfOutputs": 1,
  "highNoiseFraction": 0.8,
  "loraAdditiveScale": 0.6,
  "guidanceScalingFactor": 7.5,
  "numberOfInferenceSteps": 50
}

Output

The output of this action typically returns a list of generated image URLs. For example:

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/880412b8-1f0c-4517-988c-e6f07f8a8ff9/eef15c03-0bc2-4ca2-94c9-34eaa2f1432a.png"
]

This URL links to the artistic image generated based on the provided input parameters.

Conceptual Usage Example (Python)

Here’s how you might call the Generate Inpainting with SDXL 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 = "9ec7c313-8e61-427c-9410-430710906ca5"  # Action ID for Generate Inpainting with SDXL

# Construct the input payload based on the action's requirements
payload = {
    "width": 1024,
    "height": 1024,
    "prompt": "In the style of Vsevolod Ivanov, ancient pagan God standing in the woods and holding a laser sword",
    "refine": "no_refiner",
    "scheduler": "K_EULER",
    "applyWatermark": True,
    "negativePrompt": "",
    "promptStrength": 0.8,
    "numberOfOutputs": 1,
    "highNoiseFraction": 0.8,
    "loraAdditiveScale": 0.6,
    "guidanceScalingFactor": 7.5,
    "numberOfInferenceSteps": 50
}

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 example, the code demonstrates how to structure the input payload for the Generate Inpainting with SDXL action. Replace the placeholders with your API key and ensure the endpoint matches your setup.

Conclusion

The ferluht/sdxl-pagan Cognitive Actions, particularly the Generate Inpainting with SDXL, provide developers with a robust tool for creating artistic images tailored to specific styles and prompts. By leveraging the power of this API, you can enhance your applications with unique visual content, opening up a world of creative possibilities. Start experimenting today and transform your projects with stunning image generation capabilities!

← Back to All Articles