Spider vs Jina Reader in 2026

Both calls on one key

A team that runs Spider for crawls and Jina Reader in notebooks keeps both. Each call keeps its path, its body and its parser; the address and the key change.

# Spider, before and after
curl -X POST https://api.spider.cloud/scrape \
  -H "Authorization: Bearer $SPIDER_API_KEY" -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/pricing", "return_format": "markdown"}'
curl -X POST https://api.specstream.ai/spider/scrape \
  -H "Authorization: Bearer $SPECSTREAM_API_KEY" -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/pricing", "return_format": "markdown"}'

# Jina Reader, before and after
curl -X POST https://r.jina.ai/ -H "Authorization: Bearer $JINA_API_KEY" -d "url=$URL"
curl -X POST https://api.specstream.ai/jina -H "Authorization: Bearer $SPECSTREAM_API_KEY" -d "url=$URL"

The address form works too: https://r.jina.ai/<url> becomes https://api.specstream.ai/jina/<url>. Each page read on either route is one Extract API page at $0.0008.

What stays the same

On /spider, a scrape takes url, several by commas, return_format as markdown, text, raw, empty or a list, return_page_links and metadata, and answers a list with one object a page: url, content, status, error, costs, links and metadata. A search answers content[] with title, description and url. Every parameter is in the Spider migration guide.

On /jina, the reply starts with Title: and URL Source:, then Markdown Content:, as Jina's does. With Accept: application/json it carries code, status and data with title, url, content and publishedTime; links, images and usage.tokens are not returned. Every parameter is in the Jina Reader migration guide.

One price a page, known before the call

Spider meters each page by the gigabytes it transfers and the CPU minutes it takes, so a heavy page that needs a browser costs more than a light one. Jina meters tokens by the length of each page. On SpecStream every page costs $0.0008, whatever its size or how it was read, from a balance that never expires. A page with no text is not charged, and the reply's specstream field carries the call's price_usd.

A whole site on the same key

For a site in place of a page, send Firecrawl's /v2/map and /v2/crawl to https://api.specstream.ai/firecrawl, or a Crawl4AI server's /crawl to https://api.specstream.ai/crawl4ai, with no server to run. /crawl/stream returns one JSON result per line as pages land. A map is one page read, and each crawled page with text is one page read at $0.0008. The parameters are in the Firecrawl migration guide.

Scored on the official benchmarks

The Extract API that answers both routes scores 0.908 F1 on WCXB. The Search API on the same key scores a 97.9 F-score on all 1,000 official SimpleQA Verified questions, 978 of them correct, and 64.0 item F1 and 43.4 row F1 on all 200 official WideSearch tasks. Every score and its method is on the benchmarks page.

You decide what is kept, on each key

Create a 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}'

Search, pages and rows on one key

The same key reads up to 20 pages or documents a call with the Extract API at $0.80 per 1,000 pages, searches the live 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. Save a request and it reruns hourly to yearly, with each row labeled changed, unchanged or no longer matching.

SpecStream

The live web data engine for teams and agents. Pages at $0.80 per 1,000 through the Extract API or the Spider, Jina, Firecrawl and Crawl4AI routes, search at $5 per 1,000, and every match as structured rows at 5¢, 15¢ or $1 a row.

Firecrawl

Monthly credit plans from $19 a month for 5,000 credits, at 1 credit a page for Scrape and Crawl. SpecStream answers Firecrawl's /v2/scrape, /v2/map and /v2/crawl at https://api.specstream.ai/firecrawl, each page at $0.0008. Firecrawl vs Jina Reader.

Crawl4AI

A crawler you can host yourself, with a hosted API from $10 a month at 120 calls a minute. SpecStream answers a Crawl4AI server's /crawl and /md at https://api.specstream.ai/crawl4ai, with no server to run. Firecrawl vs Crawl4AI.

Parallel Extract

$1 per 1,000 URLs, with 600 calls a minute by default. SpecStream reads the same pages at $0.80 per 1,000 with the Extract API, up to 20 a call, and answers Parallel's search at https://api.specstream.ai/parallel. Parallel Search API compatible.