Unlocking Image Generation with Predictive Inpainting: A Guide to hvitalii/mrherman Actions

21 Apr 2025
Unlocking Image Generation with Predictive Inpainting: A Guide to hvitalii/mrherman Actions

The hvitalii/mrherman API offers developers a powerful toolset for generating images using advanced predictive inpainting methods. By leveraging pre-built Cognitive Actions, you can turn textual prompts into stunning visuals, customize image attributes, and optimize performance—all with ease. This article will detail how to use the Generate Images with Predictive Inpainting action, showcasing its capabilities and providing practical examples to streamline integration into your applications.

Prerequisites

Before you start integrating the Cognitive Actions into your application, you will need:

  • An API key for the Cognitive Actions platform.
  • Familiarity with sending HTTP requests and handling JSON payloads.

To authenticate your requests, you will typically include your API key in the request headers. This ensures secure access to the Cognitive Actions services.

Cognitive Actions Overview

Generate Images with Predictive Inpainting

Purpose

The Generate Images with Predictive Inpainting action allows you to create images from textual prompts, with optional inpainting capabilities. You can choose between two models: the schnell model for speed and the dev model for quality. This action also provides various customization options for image size, quality, and format.

Input

The action requires a prompt, with several optional parameters for customization. Below is the input schema along with an example:

{
  "prompt": "mrherman as a sparta warior",
  "loraScale": 1,
  "numOutputs": 1,
  "guidanceScale": 3.5,
  "outputQuality": 90,
  "extraLoraScale": 1,
  "inferenceModel": "dev",
  "promptStrength": 0.8,
  "imageAspectRatio": "4:5",
  "imageOutputFormat": "png",
  "numInferenceSteps": 28
}
  • Required Field:
    • prompt: A string that describes the image you want to generate.
  • Optional Fields:
    • loraScale: Scale for applying the main LoRA weights (default: 1).
    • numOutputs: Number of images to generate (default: 1).
    • guidanceScale: Scale factor for diffusion guidance (default: 3).
    • outputQuality: Quality of the output image (default: 80).
    • extraLoraScale: Scale for additional LoRA weights (default: 1).
    • inferenceModel: Select between "dev" and "schnell" models (default: "dev").
    • promptStrength: Strength of the prompt (default: 0.8).
    • imageAspectRatio: The aspect ratio for the generated image (default: "1:1").
    • imageOutputFormat: Format of the output image (default: "webp").
    • numInferenceSteps: Number of denoising steps (default: 28).

Output

The action returns a URL pointing to the generated image. Here’s an example of the output:

[
  "https://assets.cognitiveactions.com/invocations/4e0171dd-8f25-424d-bc44-b395d88024c7/e70d407d-fa54-4290-9283-816afe3b6ffb.png"
]

Conceptual Usage Example (Python)

Here’s how you could call this 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 = "5ae08d2d-4a56-472c-a438-68c4bd48bb2e" # Action ID for Generate Images with Predictive Inpainting

# Construct the input payload based on the action's requirements
payload = {
    "prompt": "mrherman as a sparta warior",
    "loraScale": 1,
    "numOutputs": 1,
    "guidanceScale": 3.5,
    "outputQuality": 90,
    "extraLoraScale": 1,
    "inferenceModel": "dev",
    "promptStrength": 0.8,
    "imageAspectRatio": "4:5",
    "imageOutputFormat": "png",
    "numInferenceSteps": 28
}

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 snippet, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The JSON payload should be structured according to the input schema for the action, ensuring that all required fields are included.

Conclusion

The Generate Images with Predictive Inpainting action within the hvitalii/mrherman API provides a robust solution for developers looking to integrate powerful image generation capabilities into their applications. With customizable parameters and the flexibility to choose between speed and quality, you are well-equipped to create engaging visuals that meet your specific needs. Explore the possibilities by implementing this action in your projects and unlock the full potential of predictive inpainting!

← Back to All Articles