Price terms as published
Each line is the vendor’s own wording, split from the price capture checked on 2026-10-14. Nothing here is rounded, converted or re-worded — the full sentences stay in the review below. Lines that carry only a price continue the plan named on the line above, in the vendor’s own order.
| # | Plan / term | Checked |
|---|---|---|
| 1 | Free | 2026-10-14 |
| 2 | $16–749 per month | 2026-10-14 |
One-liner: Firecrawl is an API that crawls, scrapes, and searches websites and returns clean Markdown or structured JSON designed to be fed directly into LLMs. It is one of the most-used “live web” data layers in LLM agent stacks.
Verified facts (checked 2026-09-02; pricing and repo stats re-read 2026-10-14)
| Attribute | Value | Source |
|---|---|---|
| GitHub stars | 191,268 | firecrawl/firecrawl via GitHub API, 2026-10-14 |
| License (core) | AGPL-3.0 | LICENSE file, verified same day |
| SDKs | MIT (e.g. firecrawl-py) |
org repo listing |
| Last push | 2026-10-13 | GitHub API (very active) |
| Created | April 2024 | GitHub API |
| Free tier | 1,000 credits/month, no credit card | official pricing page |
The AGPL/MIT split matters: using Firecrawl as a customer or developer through its API or MIT-licensed SDKs is unrestricted; modifying the core server and offering it as a service triggers AGPL obligations. Many “fully open source” write-ups skip this nuance.
How much does Firecrawl cost? Pricing, tier by tier
| Plan | Price (monthly / annual) | Credits | Concurrency |
|---|---|---|---|
| Free | $0 | 1,000/mo | 2 |
| Hobby | $19 / $16 | 5,000/mo | 5 |
| Standard | $99 / $83 | 100,000/mo | 25 |
| Growth | $399 / $333 | 500,000/mo | 50 |
| Scale | $749 / $599 | 1,000,000/mo | 100 |
Credits burn per request: Scrape and Crawl cost 1 credit per page, Search 2 credits per 10 results, browser Interact 2 credits per minute. Failed requests are nuanced: a scrape that returns no result isn’t charged, but a page that responds with an error status like 403 or 404 is still returned and costs 1 credit. Extra credits come in $5 batches with per-plan top-up rates.
Is Firecrawl free? What the free tier actually covers
Yes — Firecrawl has a genuinely usable free tier, not a 14-day trial: 1,000 credits a month, no credit card required (hosted plan). At 1 credit per page scraped or crawled, that is roughly 1,000 pages of scraping per month, and concurrency is capped at 2 concurrent requests.
Two limits worth knowing before you plan around it. First, Search costs 2 credits per 10 results, so the free tier buys fewer search queries than the raw credit count suggests. Second, browser-based Interact costs 2 credits per minute, which is the fastest way to burn the allowance; a page that needs interaction will consume a large share of a small budget. Failed requests are the one soft spot — the page notes failures are not refunded, so retries do draw down credits.
Self-hosting is a separate question from the free tier: the core is AGPL-3.0 and self-hosting costs you only server time, but you inherit the copyleft obligations. The SDKs around it (firecrawl-py and friends) are MIT, so the language bindings are not the licensing risk — the server is.
What the marketing claims don’t say
You’ll see figures like “96% content coverage” and “P95 latency of 3.4 seconds” repeated across blog roundups. Those numbers are real — they appear in Firecrawl’s own README — but they are vendor-reported benchmarks, not independent tests. Treat them as directional.
Who should use it
Search and SEO work that needs local-language pages in the index: Firecrawl returns clean markdown for pages that plain HTTP fetching hands back as scripts or boilerplate, which is what makes it useful for localized SERP audits and for feeding a RAG pipeline pages a normal fetcher cannot read. Agents and RAG pipelines that need fresh web data without building scraper infrastructure fit the same slot: the free tier covers hobby prototypes, Standard is the first seriously production-shaped tier. If you need fully local, license-clean self-hosting with no copyleft exposure, weigh the AGPL terms first — or see the scraping alternatives we track.
Pricing and repository stats verified against official sources on 2026-09-02 and re-read on 2026-10-14. This entry follows our verification policy: every claim here was checked against primary sources, not copied from other reviews.