VerdictPal
Join
← All cardsDOSSIER · Model infrastructure · SOLID · VERIFIED 2026-06-09
Firecrawl86SolidBenchmark pendingWeb extraction and agent browsing infrastructureVerified 2026-06-09

Dossier · Model infrastructure

Firecrawl

Web extraction and agent browsing infrastructure · last verified 2026-06-09

Solid
Model infrastructure
Firecrawl
86/100
ROLEWeb extraction and agent browsing infrastructure
Editorial fit
86
Source quality
74
Citation honesty
78
Privacy controls
52
Value for money
94
Speed
84
FREEFree
$16/YRHobby
$83/YRStandard
SolidVerified 2026-06-09VP·METHOD

VERDICT HISTORY

  • Firecrawl is the web-extraction infrastructure card. It should be described as context extraction for agents and research pipelines, not as a generic search engine. This pass adds current credit economics, enterprise ZDR/SOC2 claims, endpoint credit costs, and the important warning that hosted scraping is always a trust, permission, and site-terms problem.
  • Converted imported Notion research into a full flagship-ready dossier template with metrics, panes, pricing deck, scenarios, benchmark rows, and comparison slices.

EDITOR'S NOTE

Desk hands-on 2026-06: Free tier is genuinely useful; extraction quality is top-tier for research pipelines. Treat it as infrastructure, not a search toy.

AT A GLANCE

Open-source web context API for AI agents that need to search, scrape, crawl, map, monitor, interact with pages, and turn live websites into LLM-ready Markdown, screenshots, or structured JSON.

Role: Web extraction and agent browsing infrastructureCategory: Model infrastructureEditorial · hands-on

Firecrawl flagship-ready dossier: Web extraction and agent browsing infrastructure.

PUBLIC FACTS · vendor & repo

List prices and pay-as-you-go entry points we can cite without running our own bench. Each tile links to a source when possible.

Model infrastructureSetVerdictPal card ↗
synthesizedEvidenceQuality gate ↗
7Sources checkedCard sources ↗
2026-06-07Pricing checkedQuality gate ↗

Limits & product surface

Non-price vendor claims — multipliers, caps, and API scope. Detailed matrices live in subscription and SDK sections below.

Primary surface
apiCard identity ↗
Modalities
text, dataCard identity ↗
Workflow roles
[object Object], [object Object], [object Object], [object Object], [object Object]VerdictPal editorial ↗
Alternatives tracked
Exa, Tavily, Apify, Browserbase, Bright DataVerdictPal comparison set ↗

HOW IT WORKS · agent loop

The public positioning for this product — the loop we score against on VerdictPal.

Start with the wedge

Agents that need clean Markdown/JSON from arbitrary URLs

Run the representative task

Compare Firecrawl Scrape/Search/Crawl against Exa Search + Contents on five source-fetch prompts.

Check the failure modes

Credit billing climbs on crawls/batch/interact/monitor/enhanced/agent runs

Compare before recommending

Compare against Exa, Tavily, Apify before shipping advice.

METRIC LAB · 16 DIMENSIONS

Click a tile for the editorial note. Color follows score: coral, yellow, mint.

Shape

Avg 83 · 52–94

86Editorial fit

Weighted roll-up across 13 dimensions for Model infrastructure; pending desk verification if rescored from public facts.

By group

Fit89
  • Editorial fit86
  • Wedge task fit88
  • Feature depth92
Cost85
  • Free-tier utility88
  • Cost-to-value94
  • Opportunity cost72
Trust74
  • Source grounding78
  • Privacy posture52
  • Failure transparency76
  • Evidence strength74
  • Transparency92
Workflow86
  • Integration reach90
  • Setup friction84
  • Reliability82
  • Competitive position88
  • Data portability87

SEARCH MODES · 4 lenses

One input box, many retrieval postures. Filter by tier.

01Free

REST default

REST default path for Firecrawl — verify limits on the live product.

02Pro

Batch jobs

Batch jobs path for Firecrawl — verify limits on the live product.

03Pro

Streaming

Streaming path for Firecrawl — verify limits on the live product.

04Max

Webhooks

Webhooks path for Firecrawl — verify limits on the live product.

BENCHMARK LEDGER

Public rows are vendor or third-party claims we logged with a date. Desk rows are reserved for VerdictPal self-run results.

MeasureResultSource
Compare Firecrawl Scrape/Search/Crawl against Exa Search + Contents on five source-fetch promptsmarkdown cleanliness, table fidelity, citation usefulness, structured JSON quality, latency.PlannedVerdictPal benchmark plan
Crawl a docs site at 100, 1K, and 10K pagescredit burn, duplicate pages, sitemap handling, failure rate, output cleanup time.PlannedVerdictPal benchmark plan
Use Interact on three JS-heavy pagesbrowser minutes, success rate, cost, anti-bot failures, reproducibility.PlannedVerdictPal benchmark plan
Compare standard hosted, Enterprise ZDR, and self-host data pathslog retention, payload persistence, contractual clarity, support/debug trade-off.PlannedVerdictPal benchmark plan

PRICING DECK · checked 2026-06-09

Free

$0

usage

  • Free 1K credits/pages/mo
  • Verify live regional checkout before publishing procurement advice.
Desk pick

Hobby

$16

year

  • Hobby USD 16/mo annual for 5K pages
  • Verify live regional checkout before publishing procurement advice.

RAW · checked 2026-06-09: Free 1K credits/pages/mo · Hobby USD 16/mo annual for 5K pages · Standard USD 83/mo annual for 100K · Growth USD 333/mo annual for 500K · Scale USD 599/mo annual for 1M · Enterprise custom with ZDR/SSO/SLA · Scrape/Crawl/Map/Monitor 1 credit/page; Search 2 credits/10 results; Interact 2 credits/browser-minute.

RESEARCH LOG · desk notes

What we learned while building this dossier — not vendor copy.

Notion Card pipeline ↗

Notion research pass

Firecrawl is the web-extraction infrastructure card. It should be described as context extraction for agents and research pipelines, not as a generic search engine. This pass adds current credit economics, enterprise ZDR/SOC2 claims, endpoint credit costs, and the important warning that hosted scraping is always a trust, permission, and site-terms problem.

VerdictPal git ↗

Flagship-ready structure

Converted imported research into metrics, panes, scenarios, benchmark rows, comparison notes, and pricing deck.

COMPETITIVE LENS

Where Firecrawl wins for cited research — and where a rival still belongs in the stack.

Firecrawl is stronger when agents that need clean markdown/json from arbitrary urls; Exa may still win for narrower fit, procurement, or specialist depth.

Firecrawl

  • Agents that need clean Markdown/JSON from arbitrary URLs
  • Research pipelines that crawl or batch-scrape source lists

Exa

  • You only need ranked web search without crawling or page extraction
  • You need true pay-as-you-go billing instead of monthly credit plans

UNDER THE HOOD

Vendors named on the product about page — useful for procurement and privacy reviews.

web extraction
Firecrawl
crawler
Exa
structured data ingestion
Tavily
agent browsing
Apify
source cleanup
Browserbase

DEEP PANES · 6 LENSES

Editorial lenses only. Subscription and API pricing live in their own sections above.

Lens 01

Positioning

Firecrawl is framed here as Web extraction and agent browsing infrastructure. Firecrawl is the web-extraction infrastructure card. It should be described as context extraction for agents and research pipelines, not as a generic search engine. This pass adds current credit economics, enterprise ZDR/SOC2 claims, endpoint credit costs, and the important warning that hosted scraping is always a trust, permission, and site-terms problem.

Checked
2026-06-07
Evidence
synthesized
  • Agents that need clean Markdown/JSON from arbitrary URLs
  • Research pipelines that crawl or batch-scrape source lists
  • Builders who want an open-source alternative to closed web extraction APIs
  • Teams that need enterprise ZDR/SOC2/SLA language for hosted scraping

TEST SCENARIOS · hands-on lab

How we exercised the product. Step through each run before you trust the scores.

Test run

Step 1 of 4

Start from the persona in best-for item 1.

BEST FOR

  • Agents that need clean Markdown/JSON from arbitrary URLs
  • Research pipelines that crawl or batch-scrape source lists
  • Builders who want an open-source alternative to closed web extraction APIs
  • Teams that need enterprise ZDR/SOC2/SLA language for hosted scraping

AVOID IF

  • You only need ranked web search without crawling or page extraction
  • You need true pay-as-you-go billing instead of monthly credit plans
  • Your compliance bar requires EU-only processing or fully verified zero-retention terms on every tier
  • You cannot review target-site terms, robots/policies, paywalls, or anti-bot restrictions

STRENGTHS

  • Search with content
  • scrape to Markdown/HTML/screenshots/JSON
  • crawl
  • batch scrape
  • map URL discovery
  • interact browser actions
  • monitor page changes
  • Agent preview
  • SDKs/API docs
  • open-source self-host path
  • Enterprise ZDR/SSO/SLA/security

WEAKNESSES

  • Credit billing climbs on crawls/batch/interact/monitor/enhanced/agent runs
  • Credits generally do not roll over
  • Open-source and hosted cloud behavior differ
  • Enterprise ZDR must be contract-verified
  • Scraping breaks on paywalls/anti-bot/permissions/site terms
  • Some failed-agent requests may still bill

HOW IT COULD IMPROVE

  • Institutional policy friction could be reduced with a local/on-prem deployment option, or at minimum with clear data-handling documentation that compliance teams can review without an NDA.
  • Privacy posture scores 52/100, the weakest dimension on this dossier. Addressing this would meaningfully raise the editorial fit.

EDITORIAL EVIDENCE · 4 ENTRIES

TASK

Compare Firecrawl Scrape/Search/Crawl against Exa Search + Contents on five source-fetch prompts.

Pending.

↳ markdown cleanliness, table fidelity, citation usefulness, structured JSON quality, latency.

TASK

Crawl a docs site at 100, 1K, and 10K pages.

Pending.

↳ credit burn, duplicate pages, sitemap handling, failure rate, output cleanup time.

TASK

Use Interact on three JS-heavy pages.

Pending.

↳ browser minutes, success rate, cost, anti-bot failures, reproducibility.

TASK

Compare standard hosted, Enterprise ZDR, and self-host data paths.

Pending.

↳ log retention, payload persistence, contractual clarity, support/debug trade-off.

PRIVACY DEEP-DIVE · checked 2026-06-09

Firecrawl privacy materials indicate US operation/transfer and third-party providers. Enterprise materials describe zero-data retention as a plan/contract-specific feature. Treat hosted scraping as sensitive: review logs, URLs, payload retention, site permissions, and target-site terms before sending confidential corpora.

Training on your dataNot stated
EU data residencyEnterprise tier only
SOC 2 attestationNot documented
Local-first by defaultCloud only

WORKFLOW ROLES

How this tool fits into a composed research stack:

web extractioncrawlerstructured data ingestionagent browsingsource cleanup

QUALITY GATE · SOLID

Evidence ready
Benchmark pending
  • desk hands-on 2026-06-15: headline and metrics adjusted after desk trial.
  • score-revision 2026-06-09: differentiated from public facts; pending desk verification
  • Scira-gate correction 2026-06-04: git card created as source-backed solid; no public recommendation until extraction benchmark and dossierTemplate are complete.
  • Next action: Run extraction-fidelity, crawl-cost, dynamic-page, and ZDR/sensitive-source review benchmarks.
  • License: AGPL-3.0 core; SDKs/some UI components MIT; hosted cloud adds proprietary infrastructure
  • Has API: true
  • Open source: true
  • Public recommendation: false
  • evidence-override: Notion source-of-truth sync; evidence remains source-backed until desk benchmark or hands-on pass.
  • Flagship-ready structure generated from Notion deep research and imported git fields on 2026-06-09.
Editorial · hands-onSolid

Related dossiers

AI search
Exa
76/100
ROLEAI-native web retrieval and RAG source fetching

API · Text · Data

AI-native web search, contents, answer, monitor, and agent API platform for AI systems that need current web context, parsed pages, token-efficient highlights, structured outputs, and grounded citations.

Watch outYou need a consumer research UI rather than an API

Editorial · hands-onSolid
Agentic research flowsRAG pipelines that
From$7/MOSearch
Editorial fit
76
Source quality
52
Citation honesty
78
Privacy controls
57
Value for money
74
Data portability
65
Model infrastructure
OpenRouter
86/100
ROLELLM model router and cost-comparison gateway

API · Text · Code · Data

OpenAI-compatible model-routing API for comparing and shipping across many LLM providers, with pay-as-you-go credits, fallback routing, BYOK, activity logs, spend budgets, and enterprise data-policy controls.

Watch outYou need a single-vendor DPA with no subprocessors or upstream providers

Editorial · hands-onSolid
Developers comparing modelApps that need
FromCUSTOMPay-as-you-go
Editorial fit
86
Source quality
72
Citation honesty
70
Privacy controls
58
Value for money
92
Data portability
65

Browse the atlas

The Pack · Editorial newsletter

New cards in your inbox. Free.

One short email when a card ships or changes status. No tracking, no third-party analytics. Unsubscribe in one click.