Scrapfly compatible

The call, before and after

# Before
curl -G https://api.scrapfly.io/scrape \
  --data-urlencode "key=$SCRAPFLY_KEY" \
  --data-urlencode "url=https://en.wikipedia.org/wiki/Web_scraping"

# After
curl -G https://api.specstream.ai/scrapfly/scrape \
  --data-urlencode "key=$SPECSTREAM_API_KEY" \
  --data-urlencode "url=https://en.wikipedia.org/wiki/Web_scraping"

With Scrapfly's Python SDK, scrapfly-sdk 0.12.0, the change is the constructor: client = ScrapflyClient(key=SPECSTREAM_KEY, host="https://api.specstream.ai/scrapfly"). client.scrape(ScrapeConfig(url=...)) then runs as before.

What the route reads, and what it refuses

Read and applied: url, key, format with markdown or text and its options such as markdown:no_links, and proxified_response. A page the site does not have comes back as Scrapfly's 404 ERR::SCRAPE::BAD_UPSTREAM_RESPONSE, which the SDK raises as UpstreamHttpClientError.

Refused, in Scrapfly's own form: a host that does not resolve gets 422 and a wrong key gets 401, both raised by the SDK as ApiHttpClientError. Every parameter was measured on production on Nov 10, 2026, and the full pass table lists each one and its result.

Every field that comes back

The reply has uuid, config with url and format, context, and result with status, success, status_code, reason, format, content, content_type, url, size and duration. result.log_url and result.error are present and empty on a page that was read.

The route does not return result.request_headers, response_headers, cookies, screenshots, iframes, browser_data, extracted_data, dns, ssl or content_encoding.

One price a page, and a failed page is free

On Scrapfly browser rendering adds 5 credits a request, a residential proxy costs 25, and failed scrapes are billed once 30% of your traffic is blocked. On SpecStream every page read costs $0.0008, which is $0.80 per 1,000, a failed page is never charged, and a busy call gets 429 with Retry-After and charges nothing. Credit you buy never expires.

Scored on the official benchmarks

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, 978 of them correct, and 64.0 item F1 and 43.4 row F1 on all 200 WideSearch tasks. Every score and its method is on the benchmarks page.

You decide what is kept, on each key

Create the key with zero_retention and a call made with it keeps no request text and no answer after the reply. spend_cap_cents is the most the key can spend.

curl -X POST https://api.specstream.ai/v1/account/keys \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"name": "production", "scopes": ["account:read", "finds:spend"], "spend_cap_cents": 500000, "zero_retention": true}'

Pages, search and rows on one key

The same key reads up to 20 pages or documents a call with the Extract API, searches the web at $5 per 1,000 searches, and runs Deep Search, which returns every match for a request in plain words as structured rows, at 5¢, 15¢ or $1 a row, with a source URL on every value.

ScrapingBee

A page API from $19 a month for 75,000 credits, where a default call with JavaScript costs 5 credits and credits don't roll over. SpecStream answers ScrapingBee's call at https://api.specstream.ai/scrapingbee/api/v1/ too, at $0.80 per 1,000 pages.

ScraperAPI

A page API on monthly plans: Hobby is $49 a month for 100,000 API credits, and the balance resets when the subscription renews. SpecStream answers ScraperAPI's call at https://api.specstream.ai/scraperapi too, at $0.80 per 1,000 pages.

ZenRows

A page API on monthly plans from $19 for 45,000 credits, where a JavaScript-rendered request always costs 5 credits. SpecStream charges one price a page and answers ZenRows' call at https://api.specstream.ai/zenrows/v1/.

Firecrawl

Scrape, search and crawl in credits: Hobby is $19 a month for 5,000 credits. SpecStream answers Firecrawl's v2 calls at https://api.specstream.ai/firecrawl, and $19 reads 23,750 pages.