Crawl4AI
- Free allowance
- 1,000 credits every month once an email is verified, at 10 requests a minute and 2 at a time; a scrape costs 0.5 credit, so about 2,000 pages a month; 20 guest credits with no signup
- Verdict
- caveatusable, with a named limit or an unprobed read path
- Cost
- free tier
- Identity
- emailno card, no phone
- Binding limit
- request-rate10 requests a minute, 5 per 10 seconds, 2 at a time and 20 searches a minute on the free plan; 1,000 credits a month on the free plan: a page costs 0.5 credit at the base price, 1 credit in a real browser and 2 on a hard site, so the...
- Evidence
- measured18 sources; a live request was made and its output kept
- Freshness
- stale on 2027-01-02checked 2026-11-02 (live); stale on 2027-01-02
| Free tier | 1,000 credits every month once an email is verified, at 10 requests a minute and 2 at a time; a scrape costs 0.5 credit, so about 2,000 pages a month; 20 guest credits with no signup unit credits; metered on request-count; GET /v1/prices gives the free plan an allowance_per_month of 2500: a cap on all credit spent in a month, plan credits and bought extra credits together, not a second grant; the monthly grant is 1000 |
|---|---|
| Price | none stated |
| Limits |
over the 10-a-minute limit the API answers 429 and the free plan only limits the pace; at zero credit it answers 402 no_credit, and an account without a plan gets 402 plan_needed from a top-up |
| Binding | request-rate the monthly grant ends a free month before the pace does: 1,000 credits at 0.5 credit a scrape are 2,000 scrapes, which 10 requests a minute spends in 200 minutes |
| Identity | email a 7-day pass key is issued with no signup and carries the guest credits; the monthly credits need a verified email, which replaces the pass key with a permanent one; no card |
| Egress | not applicable: Markdown answers |
| Per-object cap | not applicable: the unit is a page |
| Retention | the monthly free credit resets on the 1st of each calendar month; extra credits bought on a plan never expire |
| Interface | rest REST scrape, batch and job routes; per-response cost headers |
| Tooling | REST with a Bearer key; an MCP server; x-c4-cost, x-c4-balance and x-c4-allowance-left headers on each answer |
| Composes with | Cheapest page unit in the read ladder at 0.5 credit; a plausible first leg once its free allowance is confirmed |
| Reliability | no reliability event recorded |
| Notes | GET /v1/prices declares a 1000-credit monthly grant and a 2500-credit monthly allowance cap for the free plan; at 0.5 credit a scrape the grant is 2,000 pages. x-c4-allowance-left on each answer carries what is left under the month's cap. |
| Evidence note | measured free-plan limits and grants read from GET /v1/prices with no key; the homepage and llms.txt read again |
| Parsed numbers |
|
Evidence 18
| date | source | what it proved |
|---|---|---|
| 2026-10-22 | crawl4ai.com/ | not published |
| 2026-10-30 | crawl4ai.com/ | 20 credits come with no signup and 1,000 credits a month after an email is verified |
| 2026-10-30 | probe: https://api.crawl4ai.com/v1/prices | the free plan allows 10 requests a minute, 5 per 10 seconds, 2 at a time, 20 searches a minute and 3 keys, under a 2,500-credit monthly allowance |
| 2026-10-30 | crawl4ai.com/llms.txt | the monthly allowance resets on the 1st |
| 2026-10-30 | crawl4ai.com/llms.txt | the text of Limits, above |
| 2026-10-30 | crawl4ai.com/llms.txt | a pass key comes at once; the monthly credit and a permanent key need a verified email |
| 2026-10-30 | crawl4ai.com/llms.txt | each answer carries its cost, the balance and the allowance left |
| 2026-10-30 | crawl4ai.com/llms.txt | the allowance caps plan credits and extra credits together |
| 2026-10-30 | crawl4ai.com/ | the free tier needs no card |
| 2026-10-30 | probe: https://api.crawl4ai.com/search?q=web+scraping | a keyless GET /search answered 401 with no body: the API wants a key, even a pass key |
| 2026-11-01 | crawl4ai.com/ | the free plan grants 1,000 credits every month, and a page costs 0.5 credit at the base price, 1 in a real browser and 2 on a hard site |
| 2026-11-01 | crawl4ai.com/llms.txt | POST /scrape/jobs takes up to 10,000 URLs in one submission, which drains in the background while the caller polls |
| 2026-11-01 | crawl4ai.com/llms.txt | the pace limit answers 429 and only slows the free plan, while zero credit answers 402 no_credit and a top-up needs a plan |
| 2026-11-01 | probe: https://api.crawl4ai.com/v1/prices | the live price table grants 1,000 credits a month, prices a scrape at 0.5 credit and paces the free plan at 10 requests a minute |
| 2026-11-01 | crawl4ai.com/llms.txt | a batch takes at most 50 URLs in one call and bills one unit per URL |
| 2026-11-01 | probe: https://api.crawl4ai.com/v1/prices | a scrape is 0.50 at the base level; a page a real browser rendered is 2.0 times and a hard-site page 4.0 times the base |
| 2026-11-01 | crawl4ai.com/legal/ | an account holds an email and, when one is used, a Google or GitHub sign-in identifier; card details are handled for paid tiers only |
| 2026-11-01 | crawl4ai.com/ | 20 credits come with no signup; the Free tier is $0 with no card and gives 1,000 credits every month once an email is verified, at 10 a minute and 2 at a time |
Unknown values 1
- reliability note: nothing of this kind was observed