Generating Stunning Images from Text Prompts with Llama2 Cognitive Actions

24 Apr 2025
Generating Stunning Images from Text Prompts with Llama2 Cognitive Actions

In the fast-evolving landscape of AI and machine learning, the ability to convert text descriptions into vivid images has become a highly sought-after capability. The fofr/llama2-prompter API provides developers with a powerful tool to harness this potential through its Cognitive Actions. Utilizing advanced models like Llama2, developers can create visually captivating images based on textual prompts, significantly enhancing user engagement and creativity in applications. In this article, we will explore how to integrate the Generate Image from Text Prompt action, covering its inputs, outputs, and usage examples.

Prerequisites

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

  • API Key: You will need an API key for the Cognitive Actions platform to authenticate your requests. This key should be included in the request headers.
  • Environment Setup: Make sure you have the necessary environment to make HTTP requests, such as Python with the requests library installed.

Authentication Concept

In your API calls, the authentication generally involves passing your API key in the headers as follows:

headers = {
    "Authorization": f"Bearer YOUR_COGNITIVE_ACTIONS_API_KEY",
    "Content-Type": "application/json"
}

Cognitive Actions Overview

Generate Image from Text Prompt

The Generate Image from Text Prompt action leverages the Llama2 13b base model, specially fine-tuned for generating images from text descriptions. This action is designed to enhance the process of converting textual prompts into visual representations.

Input

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

  • prompt (string, required): The textual prompt that guides the image generation.
  • seed (integer, optional): A random seed for deterministic outputs.
  • topK (integer, optional): Samples from the top K most likely tokens during decoding (default is 50).
  • topP (number, optional): Samples from the top P percentage of likely tokens (default is 0.9).
  • debug (boolean, optional): Enables additional debugging outputs (default is false).
  • temperature (number, optional): Controls the randomness of outputs (default is 0.75).
  • maxNewTokens (integer, optional): Maximum number of tokens to generate (default is 128).
  • minNewTokens (integer, optional): Minimum number of tokens to generate (default is -1; set to -1 to disable).
  • stopSequencesList (string, optional): A list of sequences where text generation should halt.
  • fineTunedWeightsPath (string, optional): Path to the fine-tuned model weights.

Example Input:

{
  "topK": 50,
  "topP": 0.9,
  "debug": false,
  "prompt": "[PROMPT] a spooky ghost, in the style of",
  "temperature": 0.75,
  "maxNewTokens": 128,
  "minNewTokens": 10,
  "stopSequencesList": "[/PROMPT]"
}

Output

The action typically returns a string that represents the generated image description, showcasing the model's creative interpretation of the prompt.

Example Output:

"kawacy, pensive surrealism, dark white and red, 32k uhd, dark orange and blue, cranberrycore, captivating"

Conceptual Usage Example (Python)

Here’s a conceptual example of how you might invoke the Generate Image from Text Prompt 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 = "5f6d4442-7a3d-4a97-934a-a0f694ab82fe" # Action ID for Generate Image from Text Prompt

# Construct the input payload based on the action's requirements
payload = {
    "topK": 50,
    "topP": 0.9,
    "debug": False,
    "prompt": "[PROMPT] a spooky ghost, in the style of",
    "temperature": 0.75,
    "maxNewTokens": 128,
    "minNewTokens": 10,
    "stopSequencesList": "[/PROMPT]"
}

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, you replace the API key and endpoint with your actual details. The payload is structured according to the required input schema, ensuring that the prompt and additional parameters are set correctly.

Conclusion

The Generate Image from Text Prompt action from the Llama2 Cognitive Actions suite empowers developers to create stunning visual content from textual descriptions effortlessly. By integrating this functionality into your applications, you can enhance user experiences and unlock new creative possibilities. Start experimenting with your text prompts today, and explore the fascinating world of AI-driven image generation!

← Back to All Articles