Harnessing Text Generation with OpenChat 3.5 AWQ Cognitive Actions

22 Apr 2025
Harnessing Text Generation with OpenChat 3.5 AWQ Cognitive Actions

Integrating advanced text generation capabilities into your applications has never been easier with the OpenChat 3.5 AWQ Cognitive Actions. This powerful API enables developers to generate diverse and contextually rich text based on user-provided prompts. By leveraging the enhanced capabilities of mixed-quality data, you can create responses that are not only coherent but also tailored to specific needs through various adjustable parameters. Whether you're developing a chatbot, content generator, or creative writing assistant, these Cognitive Actions will help streamline your workflow and enhance user experience.

Prerequisites

To get started with the OpenChat 3.5 AWQ Cognitive Actions, you'll need the following:

  • API Key: Sign up for the service and obtain your API key to authenticate requests.
  • Basic knowledge of JSON: Familiarity with JSON structures will help you construct the input payload for the actions.

The authentication process typically involves passing your API key in the request headers, enabling secure access to the action endpoints.

Cognitive Actions Overview

Generate Text with OpenChat 3.5 AWQ

The Generate Text with OpenChat 3.5 AWQ action utilizes the OpenChat 3.5 AWQ model to produce text based on the provided prompt. This action allows developers to control output diversity and randomness through adjustable parameters, making it versatile for various applications.

Input

The input for this action requires a JSON object consisting of the following fields:

  • prompt (required): A string that contains the prompt to base the text generation on.
    Example: "Write a poem about cheese. It should conclude that parmigiano reggiano is the GOAT."
  • topTokens (optional): An integer indicating the number of tokens with the highest probabilities to consider. Default is 50.
  • temperature (optional): A number used to adjust the randomness of the output. Default is 0.8.
  • promptTemplate (optional): A string that defines the template for formatting the prompt. Default is "GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:".
  • repetitionPenalty (optional): A number to discourage repeated tokens in the output. Default is 0.
  • uniquenessPenalty (optional): A number to encourage the inclusion of less frequent tokens. Default is 0.
  • probabilityThreshold (optional): A number to set a cumulative probability threshold for token selection. Default is 0.95.
  • maximumGeneratedTokens (optional): An integer indicating the maximum number of tokens to generate. Default is 512.

Example Input:

{
  "prompt": "Write a poem about cheese. It should conclude that parmigiano reggiano is the GOAT.",
  "topTokens": 50,
  "temperature": 0.8,
  "promptTemplate": "GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:",
  "repetitionPenalty": 0,
  "uniquenessPenalty": 0,
  "probabilityThreshold": 0.95,
  "maximumGeneratedTokens": 512
}

Output

The action returns a string containing the generated text based on the input prompt and parameters.

Example Output:

"In dairy lands where cows freely roam, A tale of cheese, I'm here to share, From mild to strong, each flavor toome, Parmigiano Reggiano, oh so fair. In Gouda's golden hue, a soft embrace, A velvety dance upon the tongue, From aged Cheddar's crumbly ways, To Camembert's creamy song. Blue Cheese, the brave, with veins of blue, A bold and spicy fire ignites, Brie's soft embrace, a tender hue, A symphony of flavors takes flight. In Feta's crumbly dance, a salty zest, A tangy treat, for olives and fries, Roquefort's richness, a darker quest, In every bite, a story unveiled. But in the hills of Italy's heart, A legend arises, strong and true, Parmigiano Reggiano, a work of art, A masterpiece, the pinnacle I view. In every grate, a symphony of taste, A history, a culture, a bond, From ancient times, a gift we've graced, A treasure, an icon, forever bond. For Parmigiano Reggiano, the crown is won, The GOAT of cheese, the best there ever be, A timeless classic, a love that's spun, In every mouthful, a memory to keep."

Conceptual Usage Example (Python)

Here’s a conceptual Python code snippet demonstrating how to call the Generate Text with OpenChat 3.5 AWQ 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 = "5433dc74-42f4-41ad-a985-3df9c31f3dbd" # Action ID for Generate Text with OpenChat 3.5 AWQ

# Construct the input payload based on the action's requirements
payload = {
    "prompt": "Write a poem about cheese. It should conclude that parmigiano reggiano is the GOAT.",
    "topTokens": 50,
    "temperature": 0.8,
    "promptTemplate": "GPT4 User: {prompt}<|end_of_turn|>GPT4 Assistant:",
    "repetitionPenalty": 0,
    "uniquenessPenalty": 0,
    "probabilityThreshold": 0.95,
    "maximumGeneratedTokens": 512
}

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 the Python snippet above:

  • Replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key.
  • The action_id is set to the ID for the Generate Text with OpenChat 3.5 AWQ action.
  • The payload variable is structured according to the input schema requirements.

Conclusion

The OpenChat 3.5 AWQ Cognitive Actions offer developers a powerful tool for generating rich text outputs tailored to specific prompts. By adjusting parameters such as temperature and token selection, you can fine-tune the results to meet your application's needs. Explore the potential of these actions to enhance user engagement and automate content creation in your projects. Happy coding!

← Back to All Articles