Generate Stunning Images with the fmosele/synthiola Cognitive Actions

21 Apr 2025
Generate Stunning Images with the fmosele/synthiola Cognitive Actions

In today's digital landscape, the ability to generate synthetic images has become invaluable for various applications, from creative projects to AI research. The fmosele/synthiola Cognitive Actions provide developers with a robust API for generating synthetic images using the Synthiola model within the SDXL latent space. This set of actions supports both image-to-image transformations and inpainting modes, allowing for advanced refinement and customization. By leveraging these pre-built actions, developers can easily integrate image generation capabilities into their applications, saving time and effort while enhancing creativity.

Prerequisites

To get started with the fmosele/synthiola Cognitive Actions, you will need to meet the following prerequisites:

  • API Key: You will require an API key to access the Cognitive Actions platform. This key is essential for authenticating your requests.
  • Basic Setup: Familiarity with making API calls and handling JSON data will be beneficial.

For authentication, you will typically include your API key in the headers of your requests. This allows you to securely access the various actions provided by the API.

Cognitive Actions Overview

Generate Synthetic Image with SDXL

Description: This action generates synthetic images using the Synthiola model living in the SDXL latent space. It supports img2img and inpaint modes, offering features such as mask application, refinement options, and scheduler algorithms to fine-tune image outputs.

Category: Image Generation

Input

The input schema for the action is structured as follows:

{
  "prompt": "A photo of TOK female eating a burger sloppy",
  "width": 1024,
  "height": 1024,
  "refine": "no_refiner",
  "loraScale": 1,
  "scheduler": "K_EULER",
  "guidanceScale": 7.5,
  "applyWatermark": true,
  "promptStrength": 0.8,
  "numberOfOutputs": 1,
  "highNoiseFraction": 0.8,
  "numberOfInferenceSteps": 50
}
  • prompt (string): Descriptive text guiding the image output. Example: "A photo of TOK female eating a burger sloppy".
  • width (integer): Width of the output image in pixels. Default is 1024.
  • height (integer): Height of the output image in pixels. Default is 1024.
  • refine (string): Refining technique. Options include "no_refiner", "expert_ensemble_refiner", and "base_image_refiner". Default is "no_refiner".
  • loraScale (number): Value between 0 and 1 for LoRA scaling. Default is 0.6.
  • scheduler (string): Method for scheduling the denoising process. Default is "K_EULER".
  • guidanceScale (number): Intensity of classifier-free guidance. Valid range is from 1 to 50. Default is 7.5.
  • applyWatermark (boolean): Whether to apply a watermark to the generated image. Default is true.
  • promptStrength (number): Strength of the prompt when using img2img or inpaint. Default is 0.8.
  • numberOfOutputs (integer): Number of images to output. Default is 1.
  • highNoiseFraction (number): Amount of noise in the refining process. Valid range is from 0 to 1. Default is 0.8.
  • numberOfInferenceSteps (integer): Total number of steps for denoising. Valid range is 1 to 500. Default is 50.

Output

The output of the action typically returns a URL pointing to the generated image. For example:

[
  "https://assets.cognitiveactions.com/invocations/83f4e202-8c4b-423d-863a-3fdc8e79822f/7776f705-8d9f-46bd-9ff9-afb33c9e8a69.png"
]

This URL can be used to access the generated image directly.

Conceptual Usage Example (Python)

Here’s a conceptual Python code snippet demonstrating how you might call the Generate Synthetic Image 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 = "a64f0b67-fdfe-411a-b7ee-a986cb7d8e3b"  # Action ID for Generate Synthetic Image with SDXL

# Construct the input payload based on the action's requirements
payload = {
    "width": 1024,
    "height": 1024,
    "prompt": "A photo of TOK female eating a burger sloppy",
    "refine": "no_refiner",
    "loraScale": 1,
    "scheduler": "K_EULER",
    "guidanceScale": 7.5,
    "applyWatermark": True,
    "promptStrength": 0.8,
    "numberOfOutputs": 1,
    "highNoiseFraction": 0.8,
    "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 code, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id variable is set to the ID of the Generate Synthetic Image action. The JSON payload is structured according to the input schema, and the request is sent to the hypothetical endpoint.

Conclusion

The fmosele/synthiola Cognitive Actions provide powerful capabilities for generating synthetic images, whether you are creating artwork, enhancing applications, or conducting research. By leveraging the flexibility of the input parameters and the ease of integration, developers can unlock a world of creative possibilities. Consider exploring these actions further to enhance your projects and bring your ideas to life through stunning images.

← Back to All Articles