Compare hosted alternatives to Firecrawl
Firecrawl
Proxy Compare holds 4 tiers for Firecrawl, read from firecrawl.dev and last checked 20 Aug 2026, from $19/mo. It bills in credits rather than gigabytes, so no per-GB figure is published; the per-1,000-request column below is the comparable one.
A crawl-and-scrape API billed from a monthly credit balance, where scrape, crawl, map and monitor each cost one credit per page.
Firecrawl at a glance
- Sign-up
- Unclear
- Pay with
- PayPal, card
- Refunds
- —
- Free trial
- 1,000 free credits
- IPs (their number)
- —
- Won't work for
- Nothing we know of
Plans
| Tier | Per month | Credits | Concurrency | Per 1,000 standard pages |
|---|---|---|---|---|
| HobbyPrice read 2026-09-08 | $19 | 5,000 credits/month | 5 | $3.80 |
| StandardPrice read 2026-09-08 | $99 | 100,000 credits/month | 25 | $0.99 |
| GrowthPrice read 2026-09-08 | $399 | 500,000 credits/month | 50 | $0.80 |
| ScalePrice read 2026-09-08 | $749 | 1,000,000 credits/month | 100 | $0.75 |
Monthly subscriptions, before tax. Annual discounts and extra-credit purchases are excluded. A dash means the page states a price without stating that figure beside it, not that the plan has none.
The last column is ours, not theirs: the tier price divided by its credit allowance, times the quoted Scrape credit cost. It assumes the full allowance is used. The workload calculator includes the cost of credits you buy but leave unused.
Whose network is underneath
Not established
A scraping API is a storefront over somebody’s proxy pool. This one does not say whose. We looked twice: through its pricing, about and FAQ pages, and then through the places a supplier has to be named for data-protection reasons — subprocessor, trust, security and privacy pages. Neither names a network. That is not the same as there being no upstream.
It matters because ethics do not stop at the vendor boundary. In July 2026 the FBI seized a residential proxy platform built on roughly two million consumer devices enrolled with little or no meaningful consent, and that network was resold under other brands. A clean storefront can sit on top of one.
- Billed in
- credits
- Request billing
- Returned documents are billable even for target errors such as 403 and 404. Some scans, guards, cache modes and other endpoints can also charge without a document.
Billing policy and source
What decides the charge is whether Firecrawl returned a document, not whether the target site returned a successful HTTP status code. Firecrawl returned a document: 1 credit per page, plus any of the option costs above. This includes pages that come back with an error status such as 403 Forbidden or 404 Not Found. The target responded, Firecrawl captured that response, and you get it back as a document. Check the metadata.statusCode field in the response to spot these cases and stop retrying URLs that are consistently blocked. Firecrawl returned no document: 0 credits. A scrape that fails outright, for example because the site never responded or every rendering attempt failed, is not charged. A few cases still charge when no document comes back. They cover work that Firecrawl already performed on your behalf before the scrape ended. Case What is charged Threat protection scan 2 credits per scanned URL, including a scrape that the scan itself blocked Prompt injection check 5 credits, once the checkPromptInjection guard has run FIRE-1 agent Usage-based, for the navigation the agent already did lockdown cache miss 1 credit Monitor check A page that errors during a check is charged the base 1 credit per page Interact and Browser sessions Charged per browser minute of session time, with a one-minute minimum, whether or not the session got what you wanted
Read 2026-09-09 - Concurrent requests
- 5 concurrent requests on Hobby
What one unit of work costs
A credit is not a price until you know what it buys, and the same plan charges different amounts for different work.
- Scrape
- 1 credit per request
Basic Scrape endpoint per returned document, including target error pages such as 403 and 404. Excludes PDF parsing, JSON extraction, prompt-injection checks, Zero Data Retention, threat scans, FIRE-1 navigation, lockdown cache mode and X/Twitter URLs. These options and other endpoints have separate charges, including some failures without a document.
Credit price read 2026-09-08
Questions
Does Firecrawl ask for ID to sign up?
Firecrawl does not state what it asks for at sign-up, and we have not been able to establish it from its own pages. That is a fact about the provider rather than a gap in the roster, and it is shown as one above.
How can you pay Firecrawl?
PayPal, card. No cryptocurrency option is published, so a card or an equivalent named payment is the route in. Payment methods come off the provider's own checkout and are dated with the rest of the record.
Where does Firecrawl get its addresses?
Firecrawl is a scraping API rather than a proxy network: you send it a URL and it returns the page, using infrastructure it does not itemise. There is no residential pool of its own to disclose, so this site records no sourcing model for it and does not count it among the providers that will not say.
Does Firecrawl offer a trial or a refund?
Trial: 1,000 free credits. Both are quoted from the provider's own terms rather than summarised, because a paraphrase of a refund window is the kind of precision nobody should take from us.
Keep comparing
- Firecrawl vs Crawlbase Head to head on price and policy
- Firecrawl vs Scraperapi Head to head on price and policy
- Firecrawl vs Scrapingbee Head to head on price and policy
- Providers with a free or paid trial Firecrawl qualifies; see everyone else who does
- Providers that accept PayPal Firecrawl qualifies; see everyone else who does
- What proxy vendors disclose, September 2026 9 Sept 2026
- Where Firecrawl says its addresses come from They do not say, and what that means
- Cost calculatorWhat Firecrawl actually costs at your monthly volume