Website to Markdown API in 2026

$0.80 per 1,000 pages, no plan

Every page read costs $0.0008, from the first call. Only a page returned with status ok is charged: empty, blocked, not_found, unsupported, invalid_url and failed cost nothing, and each reply carries its own price_usd. 100,000 pages cost $80.

Twenty URLs in one call

Send 1 to 20 URLs in urls. Each comes back in data.results in the order sent, with its status, text, title and dates. Each URL gets up to 30 seconds, all at once. Pages you already hold go in pages as HTML, and are converted with nothing fetched.

import os
import requests

reply = requests.post(
    "https://api.specstream.ai/v1/extract",
    headers={"Authorization": f"Bearer {os.environ['SPECSTREAM_API_KEY']}"},
    json={
        "urls": ["https://docs.python.org/3/whatsnew/3.13.html", "https://react.dev/learn"],
        "format": "markdown",
    },
    timeout=60,
).json()

for page in reply["data"]["results"]:
    if page["status"] == "ok":
        print(page["url"], page["title"])
        print(page["text"][:500])
print(reply["data"]["price_usd"])

Pages behind a refusal

A page that refuses a plain read is tried again as a browser would ask for it, then in a real browser, then from the site's own feed, then from its newest public archive capture. read_from names the one that gave the text. A bot check's page is never returned as text, and a site whose robots.txt asks us not to read it comes back blocked, unbilled.

PDF, Word and Excel in the same call

A URL that points to a PDF, Word, Excel or EPUB file is read like a web page, at the same $0.0008 a document. See PDF to Markdown API for the prices of the document tools.

Find the pages on the same key

The Search API finds the URLs first, at $1 per 1,000 on the fast search, with site:, filetype: and after: held on every result. Pass its sources to the Extract API and the model has Markdown two calls later.

The top published score

The SpecStream Extract API scores 0.908 F1 on WCXB. The Search API scores 97.9 F-score on all 1,000 SimpleQA Verified questions and 64.0 item F1 on all 200 WideSearch tasks. Every score and its method is on the benchmarks page.

Rows from the whole web on the same key

Deep Search returns every match on the live web as structured rows at 5¢, 15¢ or $1 a row, with the source URL and read time on every value and the most a run can cost approved before it starts.

Apify, Website To Markdown actor

Listed from $9.00 per 1,000 webpages, with 1,000 pages at $11.99 in its own price table. SpecStream reads 1,000 pages for $0.80.

Jina Reader

Prepend https://r.jina.ai/ to a URL and get its text, billed by the tokens in the output, with 10M free tokens on each new key. SpecStream answers the same call at https://api.specstream.ai/jina for $0.80 per 1,000 pages: see Jina Reader compatible.

Firecrawl

Scrape and crawl at 1 credit a page, with 1,000 free credits a month and Standard at $99 a month for 100,000 credits. SpecStream answers Firecrawl's own POST /v2/scrape, map and crawl at https://api.specstream.ai/firecrawl, each page one page read: see Firecrawl migration.