fshy

Home

Firecrawl

firecrawl.dev · 3 records in Compute, Crawling · JSON

Records 3

Firecrawl

Crawling

www.firecrawl.dev/pricingCrawl APIcrawler-scraper/firecrawlJSONcomparereport an error

Free allowance
500 searches or 1,000 pages scraped a month on the Free plan, 2 concurrent requests, no card
Verdict
★ qualifiesusable as stated; the read path was probed
Cost
free tier
Identity
emailno card, no phone
Binding limit
request-rate10 requests a minute on scrape, map and search; 2 a minute on crawl; 2 concurrent browsers; Hobby USD 19 a month on monthly billing, USD 16 billed annually, for 2,500 searches or 5,000 pages; Standard USD 83 a month for 50,000 or...
Evidence
documented3 sources; a first-party page was read and the sentence quoted
Freshness
stale on 2026-11-22checked 2026-10-22 (live); stale on 2026-11-22
Free tier500 searches or 1,000 pages scraped a month on the Free plan, 2 concurrent requests, no card
unit requests; metered on request-count
Pricenone stated
Limits
  • request-rate (binding) 10 requests a minute on scrape, map and search; 2 a minute on crawl; 2 concurrent browsers
  • per-object-cap crawl limit 1 to 100 pages a job; a page's Markdown is the unit
  • request-rate (binding) Hobby USD 19 a month on monthly billing, USD 16 billed annually, for 2,500 searches or 5,000 pages; Standard USD 83 a month for 50,000 or 100,000, quoted on annual billing source
the 2-a-minute crawl rate binds before the credit count at any plan size
Identityemail the Free plan states no card
Egressnot applicable: Markdown answers
Per-object capnot applicable: the unit is a page
Retention1 credit per page; a JS render or JSON extraction costs 4
Interfacerest REST v2 scrape, search, map and crawl; crawl is a job plus a poll
ToolingREST; a self-hosted build serves the same /v1/scrape shape
Composes withCrawl, read and search from one credential
ReliabilityNo failure's one crawl job
NotesIts free plan is 1,000 pages a month, which is 33 a day.
Evidence notedocumented The verdict rests on the 1,000-page monthly tier, which is documented from the pricing page. One crawl job and one pinned read were measured; the quota itself was not probed.
Parsed numbers
  • has-free-tier 1
  • request-rate 10 requests / minute
Evidence 3
datesourcewhat it proved
2026-10-22www.firecrawl.dev/pricingFree plan USD 0: 500 searches or 1,000 pages scraped a month, 2 concurrent requests, low rate limits; Hobby USD 19 a month billed monthly or USD 16 billed annually; Standard USD 83 a month on a card headed 'Prices shown are per month, billed annually'
2026-10-22source not publishedPOST with limit 3 answered 202 with a Location; the job ran then completed, 3 pages, usage credit 3, provider firecrawl, about 10 s
2026-10-22linux.do/?__provider=firecrawl200, 27,697 bytes, 742 words from a Cloudflare-walled forum in 4.5 s

www.firecrawl.dev/pricingBrowser hostcompute/firecrawlJSONcomparereport an error

Free allowance
Free plan: 1,000 credits every month, no credit card required; the allotment resets each monthly cycle and unused credits do not roll over.
Verdict
caveatusable, with a named limit or an unprobed read path
Cost
free tier
Identity
emailno card, no phone
Binding limit
request-rate1,000 credits a month on Free; one credit is one page on a basic scrape; Free plan: 10 requests a minute on /scrape, /map and /search and 2 requests a minute on /crawl, /agent and /interact; 2 concurrent browsers
Evidence
documented15 sources; a first-party page was read and the sentence quoted
Freshness
stale on 2026-12-05checked 2026-11-04 (live); stale on 2026-12-05
Free tierFree plan: 1,000 credits every month, no credit card required; the allotment resets each monthly cycle and unused credits do not roll over. Interact bills 2 credits per browser minute for code-only sessions and 7 with a prompt, with a one-minute minimum, so the free pool is 500 code-only browser minutes a month
unit requests; metered on request-count; credits are shared across search, scrape, crawl and interact; 1 credit is 1 page on a basic scrape
Pricenone stated
Limits
  • request-rate (binding) 1,000 credits a month on Free; one credit is one page on a basic scrape source
  • volume-quota Interact costs 2 credits per browser minute for code-only sessions and 7 with a prompt, with a one-minute minimum; the 1,000-credit pool is 500 code-only browser minutes a month source
  • request-rate (binding) Free plan: 10 requests a minute on /scrape, /map and /search and 2 requests a minute on /crawl, /agent and /interact; 2 concurrent browsers source
Interact sessions expire after a default TTL of 10 minutes or a default inactivity timeout of 5 minutes
Bindingrequest-rate the monthly credit pool binds before any concurrency limit, because the pool is spent by ordinary crawling
Identityemail Account created with an email address or GitHub login, with the email verified if prompted; no card required on the free plan; Search, Scrape and Interact also work with no API key through a keyless tier capped per IP address per day
Egressnot published: https://www.firecrawl.dev/pricing and https://docs.firecrawl.dev/billing price credits per call and list no data-transfer allowance or price; static egress IPs are Enterprise only
Per-object capnot stated
Retentionnot stated
Interfacerest, cli a REST API, Python and Node SDKs, a CLI (npm firecrawl-cli) and an MCP server
Toolingnot stated
Reliabilityno reliability event recorded
NotesRe-read on 2026-10-22: the free plan is 1,000 monthly credits with no card. An earlier table row called this paid-only and the browser document said it had no free tier; neither was right.
Evidence notedocumented Documented on the vendor's billing page: 1,000 credits a month, no card required, 10 requests a minute on /scrape, /map and /search and 2 on /crawl, /agent and /interact.
Parsed numbers
  • has-free-tier 1
  • request-rate 1000 credits / month
  • volume-quota 2 credits
  • request-rate 10 requests / minute
Evidence 15
datesourcewhat it proved
2026-10-22www.firecrawl.dev/pricing'1,000 credits every month, refreshed monthly, no credit card required'; 'Interact costs 2 credits per browser minute'
2026-10-28docs.firecrawl.dev/billingthe Free plan has 1,000 credits a month with no rollover; Interact costs 2 credits per minute code-only and 7 with a prompt, one-minute minimum
2026-10-28docs.firecrawl.dev/billingthe free plan costs nothing and does not need a card
2026-10-28docs.firecrawl.dev/billingthe row is a pool of metered browser minutes drawn from the monthly credits, not the size of one object
2026-10-28docs.firecrawl.dev/billingInteract bills 2 credits per minute code-only and 7 with a prompt, one-minute minimum, from a 1,000-credit monthly pool
2026-10-28docs.firecrawl.dev/billingthe Free plan is limited to 10 requests a minute on scrape, map and search, 2 on crawl, agent and interact, and 2 concurrent browsers
2026-10-28docs.firecrawl.dev/rate-limitsthe rate-limits page gives the Free plan 2 concurrent browsers and a 50,000 queued-job maximum
2026-10-28docs.firecrawl.dev/features/interactInteract sessions expire after a default 10-minute TTL or a default 5-minute inactivity timeout
2026-10-28docs.firecrawl.dev/developer-guides/workflow-automation/n8nregistration takes an email address or GitHub login, with email verification if prompted
2026-10-28docs.firecrawl.dev/rate-limitskeyless usage needs no API key and is capped per IP address per day
2026-10-28docs.firecrawl.dev/billingthe billing page states the Free plan's monthly credits and its per-minute rate limits
2026-10-28www.firecrawl.dev/pricingthe Free plan is 1,000 credits a month, about 1,000 pages, with no card; unchanged
2026-11-03www.firecrawl.dev/pricingthe only egress item is static IPs, Enterprise only; no transfer allowance or price is listed
2026-11-03docs.firecrawl.dev/rate-limitsa REST API, SDKs and a CLI exist
2026-11-03docs.firecrawl.dev/sdks/cli.mda CLI exists
More fields 4
keep rate basisno field in the source
billing modelunclassified
free tier kindallowance
burst x1.0
Unknown values 7
  • chaining: a source was read and is silent on this field
  • egress metered: egress metering was not published
  • per object cap stated: a source was read and is silent on this field
  • reliability note: nothing of this kind was observed
  • retention notes: the field may be stated in a source that was not read
  • retention stated: a source was read and is silent on this field
  • tooling: a source was read and is silent on this field

www.firecrawl.dev/Search APIcrawler-scraper/firecrawl-keylessJSONcomparereport an error

Free allowance
Free with no API key: Search, Scrape and Parse on the hosted MCP endpoint, Interact as well from official clients, the CLI, SDKs and REST API, and the research, developer and government search endpoints on Firecrawl Cloud; capped per IP address per day by a request count and a credit count...
Verdict
caveatusable, with a named limit or an unprobed read path
Cost
free tier
Identity
anonymousno card, no phone
Binding limit
request-ratetwo per-IP daily caps, a request count and a credit count, neither published; exceeding either answers 429; a keyless call from an address the vendor classes as VPN or proxy is refused, with a message naming the address as a VPN or...
Evidence
documented19 sources; a first-party page was read and the sentence quoted
Freshness
stale on 2026-12-03checked 2026-11-02 (live); stale on 2026-12-03
Free tierfree with no API key: Search, Scrape and Parse on the hosted MCP endpoint, Interact as well from official clients, the CLI, SDKs and REST API, and the research, developer and government search endpoints on Firecrawl Cloud; capped per IP address per day by a request count and a credit count, neither published
unit requests; metered on request-count; the API source code on GitHub reads the two per-IP caps from the settings KEYLESS_REQUESTS_PER_DAY and KEYLESS_CREDITS_PER_DAY and ships no default for either; the tier is off unless both are set
Pricenone stated
Limits
  • request-rate (binding) two per-IP daily caps, a request count and a credit count, neither published; exceeding either answers 429 source
  • request-rate (binding) a keyless call from an address the vendor classes as VPN or proxy is refused, with a message naming the address as a VPN or proxy and pointing to a free API key
keyless calls with no World ID credential are accepted from IPv4 client addresses only; a call from an IPv6 address falls through to HTTP 401
Bindingrequest-rate the per-IP daily request and credit caps bind first, and their numbers are not published
Identityanonymous no account and no API key
Egressnot applicable
Retentioneach per-IP request counter and credit counter expires 86400 seconds after it is created
Interfacerest REST at api.firecrawl.dev with no key (the Research Index at GET /v2/search/research/papers, Search, Scrape and Interact) and the hosted MCP keyless endpoint
Toolingplain HTTP with no Authorization header, an SDK, the CLI run as npx -y firecrawl-cli@latest, or an MCP client pointed at https://mcp.firecrawl.dev/v2/mcp
Composes withThe works index's last leg, after OpenAlex, TinyFish and Exa
Reliabilitya keyless call from a VPN or proxy address is refused outright, so the route depends on the class of the egress address
NotesTwo caps and no number is worse than one cap with a number: there is no threshold to design against.
Evidence notedocumented free tier and caps quoted from the vendor's rate-limit page; the refusal of a VPN or proxy address measured with a keyless probe
Parsed numbers
  • has-free-tier 1
Evidence 19
datesourcewhat it proved
2026-10-22source not publishedcaps works, two keyless-ip pools, limit null: the cap is not published
2026-10-30docs.firecrawl.dev/rate-limitskeyless use is free, capped per IP address per day by a request count and a credit count, neither number published; Search, Scrape and Parse on the hosted MCP, Interact from official clients and REST, and the research search endpoints are keyless
2026-10-30docs.firecrawl.dev/rate-limitsthe research search endpoints are keyless on Firecrawl Cloud
2026-10-30www.firecrawl.dev/blog/firecrawl-keyless-launchthe keyless caps reset every day
2026-10-30probe: https://api.firecrawl.dev/v2/search/research/papers?query=attention%20is%20all%20you%20need&k=3a keyless call from an address the vendor classes as VPN or proxy is refused, with a pointer to a free API key
2026-10-30www.firecrawl.dev/blog/firecrawl-keyless-launchkeyless use needs no account and no key
2026-10-30raw.githubusercontent.com/firecrawl/firecrawl/main/apps/api/the server refuses keyless calls from VPN, proxy or TOR addresses when its address-reputation check is configured
2026-10-30probe: https://api.firecrawl.dev/v2/search/research/papers?query=attention%20is%20all%20you%20need&k=3the refusal of the keyless research search from a VPN or proxy address is HTTP 403
2026-11-01raw.githubusercontent.com/firecrawl/firecrawl/main/apps/api/the API source refuses a keyless call from a client address that is not IPv4 with the 401 answer, unless the caller presents a World ID credential
2026-11-01www.firecrawl.dev/agent-onboarding/SKILL.mdthe vendor lists the clients that work with no key (MCP, CLI, SDK, plain HTTP), the MCP endpoint and the CLI command, and says a call from some networks (VPNs, proxies) is refused
2026-11-01raw.githubusercontent.com/firecrawl/firecrawl/main/apps/api/the API source expires each per-IP request counter and credit counter 86400 seconds after it is created
2026-11-01raw.githubusercontent.com/firecrawl/firecrawl/main/apps/api/the API source takes the two per-IP caps from settings that have no default, and the keyless tier is off unless both are configured
2026-11-01probe: https://api.firecrawl.dev/v2/search/research/papers?query=attention%20is%20all%20you%20need&k=3a keyless GET of the research papers search, sent from one egress address with no proxy, answered HTTP 403 on 2026-11-01, as it did on 2026-10-30
2026-11-01api.firecrawl.dev/v2/search/research/papers?query=attention%a keyless GET of the research papers search answered HTTP 200 with success true and three results on 2026-11-01 from a Cloudflare Workers fetch egress, an address the vendor does not refuse
2026-11-01api.firecrawl.dev/v2/search/research/papers?query=retrieval%a keyless GET of the research papers search with a second query also answered HTTP 200 with success true and two results on 2026-11-01 from the same kind of egress
2026-11-01raw.githubusercontent.com/firecrawl/firecrawl/main/apps/api/the vendor's API answers HTTP 403 with a message naming the address as a VPN or proxy and pointing to a free API key when its address check flags a keyless caller
2026-11-01www.firecrawl.dev/blog/firecrawl-keyless-launchkeyless use needs no account and no API key
2026-11-01xitter.cf/ericciarla/status/2093375835679977570a public post announcing Firecrawl keyless states that the keyless credit and rate limits vary based on demand (reported, from a mirror of the post)
2026-11-01docs.firecrawl.dev/rate-limitskeyless use is free and capped per IP address per day by a request count and a credit count, and the page states no number for either
Unknown values 2
  • egress terms: the field does not apply to this kind of offer
  • per object cap stated: the field does not apply to this kind of offer