Refinery — HTML to LLM Text (Cut RAG Token Cost)
Pricing
from $2.00 / 1,000 html extractions
HTML to LLM text cleaner for RAG pipelines. Strip scripts, nav & layout junk after Firecrawl or your fetch. BeautifulSoup-alternative speed. $0.002/page · 3 README demos in Console.
You can access the Refinery — HTML to LLM Text (Cut RAG Token Cost) programmatically from your own applications by using the Apify API. You can also choose the language preference from below. To use the Apify API, you’ll need an Apify account and your API token, found in Integrations settings in Apify Console.
1import { ApifyClient } from 'apify-client';2
3// Initialize the ApifyClient with your Apify API token4// Replace the '<YOUR_API_TOKEN>' with your token5const client = new ApifyClient({6 token: '<YOUR_API_TOKEN>',7});8
9// Prepare Actor input10const input = {11 "urls": [12 "https://example.com"13 ],14 "raw_payload": "<html><head><script>gtag('event','page_view')</script><style>.nav,.footer{display:block}</style></head><body><nav>Home · Pricing · Login</nav><article><h1>Quarterly update</h1><p>We reduced embedding cost by cleaning HTML before chunking.</p></article><footer>© 2026 · Legal · Careers</footer></body></html>"15};16
17// Run the Actor and wait for it to finish18const run = await client.actor("larelabs/refinery-html-to-llm-cleaner").call(input);19
20// Fetch and print Actor results from the run's dataset (if any)21console.log('Results from dataset');22console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);23const { items } = await client.dataset(run.defaultDatasetId).listItems();24items.forEach((item) => {25 console.dir(item);26});27
28// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docsThe Apify API client for JavaScript is the official library that allows you to use Refinery — HTML to LLM Text (Cut RAG Token Cost) API in JavaScript or TypeScript, providing convenience functions and automatic retries on errors.
Install the apify-client
$npm install apify-clientOther API clients include: