Already on Firecrawl? Reuse the same core endpoints.
SuperScraper exposes the same core paths for scrape, crawl, map, extract, batch, screenshot, and search. Point REST calls at us with a SuperScraper key. Below is the path map and a ground-truthed comparison. Where we are thinner, we say so.
First call
curl -X POST https://api.superscraper.dev/v1/scrape \
-H "Authorization: Bearer $SUPERSCRAPER_KEY" \
-H "Content-Type: application/json" \
-d '{"url":"https://example.com"}'Not a drop-in of the Firecrawl SDK. Core request paths align; response fields may differ — spot-check once (including _confidence and job payloads). Map crawl status to GET /v1/jobs/:id.
Endpoint mapping
| Firecrawl | SuperScraper |
|---|---|
| POST /v1/scrape | POST /v1/scrape(similar formats & actions) |
| POST /v1/crawl | POST /v1/crawl(include/exclude paths + webhook) |
| GET /v1/crawl/:id | GET /v1/jobs/:id(job status path differs) |
| POST /v1/map | POST /v1/map |
| POST /v1/extract | POST /v1/extract(schema-based) |
| POST /v1/batch/scrape | POST /v1/batch |
| screenshot (format) | POST /v1/screenshot(also a dedicated endpoint) |
| POST /v1/search | POST /v1/search |
Core endpoints
| Capability | SuperScraper | Firecrawl |
|---|---|---|
| Scrape (markdown / HTML / links / JSON / screenshot) | ||
| Crawl with include/exclude paths + webhook | ||
| Map / URL discovery | ||
| Batch (sync ≤100 + async) | ||
| Schema-based extract (single + multi-URL) | ||
| Dedicated screenshot endpoint(Firecrawl ships it as a format) | ||
| Browser actions (click/scroll/write/press/wait) | ||
| Search + scrape |
Output & extraction
| Capability | SuperScraper | Firecrawl |
|---|---|---|
| Clean, LLM-ready markdown | ||
| Content shaping (main-only / include-exclude tags) | ||
| Free JSON-LD / Open Graph extraction(where a site ships structured data) | ||
| Confidence score on extracted JSON | ||
| llms.txt generation |
Operations
| Capability | SuperScraper | Firecrawl |
|---|---|---|
| Automatic stealth-browser escalation | ||
| Proxy rotation handled for you | ||
| Per-domain cheapest-path cache | ||
| Never billed for failed scrapes |
Distribution
| Capability | SuperScraper | Firecrawl |
|---|---|---|
| Node + Python SDKs(in progress, not published) | ||
| CLI(in progress, not published) | ||
| Plain REST + bearer auth | ||
| MCP server | ||
| Compatible core request paths(not field-for-field SDK parity) | ||
| Long-tail language SDKs (Go/Rust/Java/…)(on our backlog) |
Performance
Measured 2026-06-23| Metric | SuperScraper | Firecrawl |
|---|---|---|
| Median scrape latency (p50) | 2.17s | we have not measured this |
| Anti-bot pass rate, 27 live sites | 96.3% (26/27) | we have not measured this |
| Paid-proxy spend on that run | $0 | we have not measured this |
Our numbers come from one dated run against 27 live sites on 2026-06-23. The script and the raw per-site rows are published, so you can re-run it: see the method and what it does not prove. The Firecrawl column is empty because we have not benchmarked their stack. We will not put a number there until we do.
Firecrawl is a third-party trademark. SuperScraper is not affiliated with or endorsed by Firecrawl.