Skip to main content

​
Log Perplexity requests

Monitor your Perplexity API calls by logging requests and responses asynchronously. Track metrics like cost, duration, and performance for Perplexity’s search-augmented models. 
import requests

url = "https://api.keywordsai.co/api/request-logs/create/"
payload = {
    "model": "llama-3.1-sonar-small-128k-online",
    "prompt_messages": [
        {
            "role": "user",
            "content": "What are the latest developments in AI?"
        }
    ],
    "completion_message": {
        "role": "assistant",
        "content": "Recent AI developments include advances in large language models..."
    },
    "cost": 0.0008,
    "generation_time": 4.2,
    "customer_params": {
        "customer_identifier": "user_404"
    }
}
headers = {
    "Authorization": "Bearer YOUR_KEYWORDS_AI_API_KEY",
    "Content-Type": "application/json"
}

response = requests.post(url, headers=headers, json=payload)

Get Started with Logging

Learn how to set up comprehensive logging for all your LLM requests