ScrapeGraphAI

ScrapeGraphAI is an open-source Python library that uses LLMs and direct graph logic to extract web data by describing what you want in plain English.

✓ JS Rendering✓ CAPTCHA BypassResidential Proxies✓ Free Tier

🚀 Visit Official Site ← Back to Directory

📊 Pros & Cons

✅ Pros

▶Prompt-driven extraction: specify the target output schema in plain English
▶100% open-source and free to self-host with zero software licensing costs
▶Compatible with OpenAI, Google Gemini, Anthropic, Azure, and local Ollama models

⚠️ Cons

▶Token consumption costs can grow rapidly when processing large batches of HTML through LLMs
▶Speed is dependent on the latency and throughput of the underlying LLM provider

🎯 Best For

Data scientists and Python engineers who want to extract structured web data using LLMs without maintaining CSS selectors or XPath rules.

🐍 Python Code Example

python

from scrapegraphai.graphs import SmartScraperGraph

graph_config = {
    'llm': {'model': 'openai/gpt-4o-mini', 'api_key': 'YOUR_OPENAI_KEY'},
    'headless': True
}

smart_scraper = SmartScraperGraph(
    prompt='Extrae los titulos y precios de los productos',
    source='https://example.com/store',
    config=graph_config
)
result = smart_scraper.run()
print(result)

🧩 Supported SDKs & Languages

PythonREST API Integration