text-generation

Generate High-Quality Text Effortlessly with Starling

27 Apr 2025
Generate High-Quality Text Effortlessly with Starling

In the world of content creation and text generation, the quality and relevance of generated text are paramount. Enter Starling, a powerful tool that leverages the advanced capabilities of the Starling-LM-7B-alpha language model. Fine-tuned from Openchat 3.5 using Reinforcement Learning from AI Feedback (RLAIF), this service enables developers to generate nuanced and contextually appropriate text outputs with remarkable speed and precision. Whether you are building chatbots, content generation applications, or educational tools, Starling can significantly enhance your projects by providing high-quality text generation capabilities.

Prerequisites

To make use of Starling, you will need a Cognitive Actions API key and a basic understanding of making API calls.

Generate Text Using Starling-LM-7B-alpha

The Generate Text Using Starling-LM-7B-alpha action is designed to produce high-quality text based on the input prompt you provide. This action solves the problem of generating coherent and contextually relevant text, making it an ideal choice for a variety of applications.

Input Requirements:

  • Prompt: The initial text input that guides the model's response (required).
  • Top K: An integer that defines the number of highest probability vocabulary tokens considered during sampling (default is 50).
  • Top P: A numeric value that applies nucleus sampling, focusing on the smallest set of top probability tokens to meet a cumulative probability (default is 0.95).
  • Temperature: A number that controls the randomness of the output, with lower values yielding more deterministic results (default is 0.2).
  • Max New Tokens: The maximum number of tokens the model should generate as output (default is 512).
  • Prompt Template: A template for the prompt containing placeholders for dynamic content (default structure includes "GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant:").

Example Input:

{
  "topK": 50,
  "topP": 0.95,
  "prompt": "Who was Henry Kissinger?",
  "temperature": 0.2,
  "maxNewTokens": 512,
  "promptTemplate": "GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant:"
}

Expected Output: The output will be a coherent and contextually relevant text about Henry Kissinger, potentially including details about his life, career, and impact on American foreign policy.

Use Cases for this Specific Action:

  • Chatbots and Virtual Assistants: Enhance user interactions by providing informative and engaging responses.
  • Content Creation: Automatically generate articles, summaries, or reports based on user-defined topics.
  • Educational Tools: Provide detailed explanations or answers to questions in an interactive learning environment.
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "bc169ce7-4a47-441d-8b47-3bcfcb47ddc2" # Action ID for: Generate Text Using Starling-LM-7B-alpha

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "topK": 50,
  "topP": 0.95,
  "prompt": "Who was Henry Kissinger?",
  "temperature": 0.2,
  "maxNewTokens": 512,
  "promptTemplate": "GPT4 Correct User: {prompt}<|end_of_turn|>GPT4 Correct Assistant:"
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    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 (non-JSON): {e.response.text}")
    print("------------------------------------------------")

Conclusion

Starling offers developers an incredible opportunity to integrate high-quality text generation into their applications. With its advanced language model and flexible input parameters, you can create dynamic and contextually appropriate content that improves user engagement and satisfaction. Start exploring the potential of Starling today, and take your projects to the next level!

← Back to All Articles