Generate Stunning Images with matanga1-2/matang-flux Cognitive Actions

22 Apr 2025
Generate Stunning Images with matanga1-2/matang-flux Cognitive Actions

In the evolving landscape of artificial intelligence, image generation and manipulation have become vital tools for developers. The matanga1-2/matang-flux spec provides a powerful Cognitive Action that allows you to generate images using advanced techniques like inpainting and image-to-image translation. With customizable parameters, developers can create high-quality images tailored to specific needs. In this article, we’ll explore the capabilities of the Generate Image with Inpainting and Translation action and guide you through integrating it into your applications.

Prerequisites

Before you start using Cognitive Actions, ensure you have the following:

  • An API key for accessing the Cognitive Actions platform.
  • Basic knowledge of JSON and how to make HTTP requests.

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

Cognitive Actions Overview

Generate Image with Inpainting and Translation

The Generate Image with Inpainting and Translation action enables developers to create images based on customizable parameters. This action supports both image-to-image translation and inpainting, providing flexibility in how images are generated. You can select between two models for different processing speeds: the dev model for quality and the schnell model for faster outputs.

Input

The input for this action requires a JSON object that includes the following fields:

  • prompt (required): A description that guides the image generation.
  • mask (optional): URI of the image mask used for inpainting.
  • seed (optional): An integer seed for reproducible results.
  • image (optional): URI of the input image for processing.
  • model (optional): Choose between dev and schnell (default is dev).
  • width and height (optional): Dimensions of the output image, applicable only in custom aspect ratios.
  • speedMode (optional): Enable fast predictions (default is false).
  • megapixels (optional): Approximate megapixels for the output image.
  • guidanceScale (optional): Influence of the prompt on the image generation.
  • ... other optional parameters

Example Input:

{
  "model": "dev",
  "prompt": "a cartoonish version of MATANG , a 33-year old asian Yaman male from Israel, after he lost 10 pound, with military-style buzzcut hair, sitting anxious against his Macbook coding in evening.",
  "speedMode": false,
  "megapixels": "1",
  "guidanceScale": 3,
  "mainLoraScale": 1,
  "promptStrength": 0.8,
  "numberOfOutputs": 1,
  "outputImageFormat": "webp",
  "aspectRatioSetting": "1:1",
  "imageOutputQuality": 80,
  "additionalLoraScale": 1,
  "numberOfInferenceSteps": 28
}

Output

The action returns a URL to the generated image in the specified format. For instance, the output could look like this:

Example Output:

[
  "https://assets.cognitiveactions.com/invocations/8b881f70-0fd1-4c38-903d-c74ebeecef1e/b226dced-3c7a-4d22-87ac-ff53236d02e5.webp"
]

Conceptual Usage Example (Python)

Here’s how you might call this action using Python, structuring the input correctly:

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 = "322d7da6-4ab0-40c9-a08b-9a73250bf60a"  # Action ID for Generate Image with Inpainting and Translation

# Construct the input payload based on the action's requirements
payload = {
    "model": "dev",
    "prompt": "a cartoonish version of MATANG , a 33-year old asian Yaman male from Israel, after he lost 10 pound, with military-style buzzcut hair, sitting anxious against his Macbook coding in evening.",
    "speedMode": False,
    "megapixels": "1",
    "guidanceScale": 3,
    "mainLoraScale": 1,
    "promptStrength": 0.8,
    "numberOfOutputs": 1,
    "outputImageFormat": "webp",
    "aspectRatioSetting": "1:1",
    "imageOutputQuality": 80,
    "additionalLoraScale": 1,
    "numberOfInferenceSteps": 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 code snippet, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id corresponds to the specific action you wish to execute. The payload is structured according to the input schema for the action.

Conclusion

The Generate Image with Inpainting and Translation action from the matanga1-2/matang-flux spec opens up exciting possibilities for developers looking to create and customize images programmatically. By leveraging the power of customizable parameters, you can produce high-quality images tailored to your project's needs. Explore further by experimenting with different prompts and settings, and consider how these capabilities can enhance your applications!

← Back to All Articles