Unlocking Creativity: Integrate Image Generation with the dgtlcorp/tero Cognitive Actions

21 Apr 2025
Unlocking Creativity: Integrate Image Generation with the dgtlcorp/tero Cognitive Actions

In the world of digital content creation, the ability to generate high-quality images on demand can significantly enhance user experience and engagement. The dgtlcorp/tero Cognitive Actions provide developers with powerful tools to create images based on detailed prompts, including inpainting capabilities and customization options through LoRA weights. This guide will walk you through the process of integrating the Generate Image with Inpainting and LoRA Customization action into your applications, enabling you to harness the full potential of image generation.

Prerequisites

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

  • An API key for the Cognitive Actions platform to authenticate your requests.
  • Basic knowledge of JSON and how to work with APIs.
  • A Python environment set up to test the integration.

Authentication typically involves including an API key in the request headers, allowing you to securely access the Cognitive Actions.

Cognitive Actions Overview

Generate Image with Inpainting and LoRA Customization

This action generates images based on specific input parameters, including inpainting options, custom LoRA weights, and various image settings such as aspect ratio, resolution, and output format. It is optimized for both quality and speed, offering two models for different needs: 'dev' and 'schnell'.

Input Schema:

The input for this action is a composite request object. Below is a breakdown of the required and optional fields:

  • Required:
    • prompt: (string) The text prompt guiding the image generation.
  • Optional:
    • mask: (string) URI of the image mask for inpainting mode.
    • seed: (integer) Seed for reproducibility.
    • image: (string) URI of an input image for inpainting.
    • width: (integer) Width of the generated image (256 to 1440 pixels).
    • height: (integer) Height of the generated image (256 to 1440 pixels).
    • goFast: (boolean) Enable fast predictions.
    • imageAspectRatio: (string) Aspect ratio for the generated image.
    • imageOutputFormat: (string) Format of the output images (webp, jpg, png).
    • numOutputs: (integer) Number of output images (1-4).
    • guidanceScale: (number) Scale of guidance during the diffusion process.
    • outputQuality: (integer) Quality of output images (0-100).
    • numInferenceSteps: (integer) Number of denoising steps (1-50).

Example Input:

{
  "goFast": false,
  "prompt": "A cozy winter coffee shop setting with warm, inviting lighting and rustic wooden interiors...",
  "loraScale": 1,
  "modelType": "dev",
  "megapixels": "1",
  "numOutputs": 4,
  "guidanceScale": 3,
  "outputQuality": 80,
  "extraLoraScale": 1,
  "promptStrength": 0.8,
  "imageAspectRatio": "1:1",
  "imageOutputFormat": "webp",
  "numInferenceSteps": 28
}

Output:

The action returns URLs to the generated images. For example:

[
  "https://assets.cognitiveactions.com/invocations/af6f9750-08d8-4ff7-bad8-e67b89bc4218/e68ad662-12f3-4d84-bf3e-96c4d70c9cb1.webp",
  "https://assets.cognitiveactions.com/invocations/af6f9750-08d8-4ff7-bad8-e67b89bc4218/7418676b-fd7b-45dd-8123-2cdbce21664c.webp"
]

Conceptual Usage Example (Python):

Here's how you might implement this action in 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 = "4e2f1a72-046e-4f0b-a752-a574cbf9fcf5"  # Action ID for Generate Image with Inpainting and LoRA Customization

# Construct the input payload based on the action's requirements
payload = {
    "goFast": False,
    "prompt": "A cozy winter coffee shop setting with warm, inviting lighting and rustic wooden interiors...",
    "loraScale": 1,
    "modelType": "dev",
    "megapixels": "1",
    "numOutputs": 4,
    "guidanceScale": 3,
    "outputQuality": 80,
    "extraLoraScale": 1,
    "promptStrength": 0.8,
    "imageAspectRatio": "1:1",
    "imageOutputFormat": "webp",
    "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 code snippet, replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id variable contains the ID for the image generation action. The payload variable is constructed to match the required input schema, which is then sent in the POST request to the hypothetical execution endpoint.

Conclusion

The dgtlcorp/tero Cognitive Actions provide a robust way to generate images tailored to specific requirements, enhancing the creative capabilities of your applications. By following the guidelines and examples outlined in this article, you can seamlessly integrate image generation into your projects. Consider exploring further customization options, experimenting with different prompts, and utilizing the inpainting feature to create unique visual content that captivates your audience. Happy coding!

← Back to All Articles