Skip to main content
For websites with changing structures or one-off extraction needs, Olostep offers LLM-powered extraction. This approach:
  • Feeds the content to a Large Language Model
  • Instructs the model to parse and return only the specified data
  • Returns a clean JSON structure containing exactly what you need
import requests
import json

def extract_with_llm():
    url = "https://api.olostep.com/v1/scrapes"

    headers = {
        "Authorization": "Bearer <API_KEY>",
        "Content-Type": "application/json"
    }

    data = {
        "url_to_scrape": "https://www.berklee.edu/events/stefano-marchese-friends",
        "formats": [
            "markdown", 
            "json"
        ],
        "llm_extract": {
            "schema": {
                "event": {
                    "type": "object",
                    "properties": {
                        "title": {"type": "string"},
                        "date": {"type": "string"},
                        "description": {"type": "string"},
                        "venue": {"type": "string"},
                        "address": {"type": "string"},
                        "start_time": {"type": "string"}
                    }
                }
            }
        },
    }

    response = requests.post(url, headers=headers, json=data)
    result = response.json()

    # The LLM extract will be available in the result
    print(json.dumps(result, indent=2))

    return result

if __name__ == "__main__":
    extract_with_llm()
You can either pass the schema or a prompt to the LLM. 

import requests
import json

def extract_with_llm():
    url = "https://api.olostep.com/v1/scrapes"

    headers = {
        "Authorization": "Bearer <API_KEY>",
        "Content-Type": "application/json"
    }

    data = {
        "url_to_scrape": "https://www.berklee.edu/events/stefano-marchese-friends",
        "formats": [
            "markdown", 
            "json"
        ],
        "llm_extract": {
            "prompt": "Extract the event title, date, description, venue, address, and start time from the event page."
        },
    }

    response = requests.post(url, headers=headers, json=data)
    result = response.json()

    # The LLM extract will be available in the result
    print(json.dumps(result, indent=2))

    return result

if __name__ == "__main__":
    extract_with_llm()
The prompt is a natural language string that is passed to the LLM to extract the data. The LLM decides how to extract the data based on the prompt. You can use this when you don’t want to use a schema. Sample Response: 
{
    "id": "scrape_94iqy385ty",
    ...
    "result": {
        "json_content":  "{\"event\":{\"title\":\"Stefano Marchese and Friends\",\"date\":\"Wednesday / January 22, 2025\",\"description\":\"Join acclaimed Italian singer-songwriter and educator Stefano Marchese for an unforgettable evening of musical magic as he takes the stage alongside a constellation of extraordinary talent in a concert titled Concerto di Duetti.\",\"venue\":\"David Friend Recital Hall (DFRH)\",\"address\":\"921 Boylston Street Boston MA 02115 United States\",\"start_time\":\"7:30 p.m. (EST)\"}}"
    }
}
json_content is the stringified JSON content of the event. You can access it as a JSON object by parsing the string. 
import json

event = json.loads(result["json_content"])
print(event["event"]["title"])
⌘I