Generate Stunning Images with the pimentoml/helmut Cognitive Actions

24 Apr 2025
Generate Stunning Images with the pimentoml/helmut Cognitive Actions

In the world of artificial intelligence, image generation has gained tremendous popularity, allowing developers to create breathtaking visuals with ease. The pimentoml/helmut spec provides powerful Cognitive Actions that leverage advanced techniques to generate and refine images. These pre-built actions offer flexibility and customization, enabling developers to integrate image generation capabilities into their applications seamlessly.

Prerequisites

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

  • API Key: You will need an API key to authenticate your requests to the Cognitive Actions platform. This will typically be passed in the headers of your requests.
  • Basic Understanding of JSON: Since the input and output are structured as JSON, familiarity with JSON formatting will help you craft your requests and handle responses effectively.

Cognitive Actions Overview

Generate Image with Inpainting and Refinement

Description: This action generates images using inpainting techniques, allowing for refinement through advanced schedulers and guidance prompts. It supports various refinement styles and adjustable parameters for image size, quality, and watermark application.

Category: Image Generation

Input

The input requires a composite request object with the following fields:

  • mask (string): URI of the input mask for the inpaint mode. Black areas will be preserved, while white areas will be inpainted.
  • seed (integer, optional): Random seed for variability. Leave blank for a new random seed each time.
  • image (string): URI of the input image for img2img or inpaint modes.
  • 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"): Text prompt to guide image generation.
  • refine (string, default: "no_refiner"): Specifies the refinement style.
  • loraScale (number, default: 0.6): Adjustment scale for LoRA.
  • scheduler (string, default: "K_EULER"): Scheduler for guiding the denoising process.
  • guidanceScale (number, default: 7.5): Scaler for classifier-free guidance.
  • applyWatermark (boolean, default: true): Determines if a watermark is applied.
  • negativePrompt (string, default: ""): Text prompt to avoid certain elements in image generation.
  • promptStrength (number, default: 0.8): Intensity of the prompt's influence.
  • numberOfOutputs (integer, default: 1): Number of images to generate (1-4).
  • refinementSteps (integer, optional): Number of refinement steps for 'base_image_refiner'.
  • highNoiseFraction (number, default: 0.8): Fraction of noise for 'expert_ensemble_refiner'.
  • numberOfInferenceSteps (integer, default: 50): Total denoising steps to perform (1-500).

Example Input:

{
  "width": 1024,
  "height": 1024,
  "prompt": "white woman in blue dress in the style of TOK",
  "refine": "no_refiner",
  "loraScale": 0.6,
  "scheduler": "K_EULER",
  "guidanceScale": 7.5,
  "applyWatermark": true,
  "negativePrompt": "",
  "promptStrength": 0.8,
  "numberOfOutputs": 4,
  "highNoiseFraction": 0.8,
  "numberOfInferenceSteps": 30
}

Output

The action typically returns an array of image URLs corresponding to the generated images.

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/1e60528e-00ca-43ee-93de-2b509a8c00bb/71c6e0a7-c91a-47e0-82da-95ac803fb003.png",
  "https://assets.cognitiveactions.com/invocations/1e60528e-00ca-43ee-93de-2b509a8c00bb/cefe2c04-f419-484f-a900-c26dbe69f8c2.png",
  "https://assets.cognitiveactions.com/invocations/1e60528e-00ca-43ee-93de-2b509a8c00bb/5a167fb0-3acd-4709-b43b-122de0393154.png",
  "https://assets.cognitiveactions.com/invocations/1e60528e-00ca-43ee-93de-2b509a8c00bb/5afdb385-98a1-429e-8639-0c4eb2dcb60b.png"
]

Conceptual Usage Example (Python)

Here's a conceptual Python code snippet to demonstrate how you might call this action:

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 = "5dab5b3f-f49a-42b8-baf1-5945a5d887d3" # Action ID for Generate Image with Inpainting and Refinement

# Construct the input payload based on the action's requirements
payload = {
    "width": 1024,
    "height": 1024,
    "prompt": "white woman in blue dress in the style of TOK",
    "refine": "no_refiner",
    "loraScale": 0.6,
    "scheduler": "K_EULER",
    "guidanceScale": 7.5,
    "applyWatermark": True,
    "negativePrompt": "",
    "promptStrength": 0.8,
    "numberOfOutputs": 4,
    "highNoiseFraction": 0.8,
    "numberOfInferenceSteps": 30
}

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 is set to the ID of the "Generate Image with Inpainting and Refinement" action.
  • The payload is constructed according to the defined input schema, demonstrating how to set various parameters.

Conclusion

The pimentoml/helmut Cognitive Actions provide robust tools for developers looking to integrate image generation and refinement into their applications. By leveraging the capabilities of actions like "Generate Image with Inpainting and Refinement", you can create stunning visuals tailored to your specifications. Explore these actions and consider how they can enhance your projects, from content creation to interactive applications. Happy coding!

← Back to All Articles