Home / Terms / Headless browser
What is a headless browser?
A headless browser is a web browser controlled by software without displaying a visible window to a person. It still loads pages, runs JavaScript, stores cookies, and can follow scripted interactions; “headless” describes how the browser is operated, while “cloud browser” says where someone else hosts it. (Browserless unit consumption, Cloudflare Browser Rendering, and Browserbase pricing, checked 2026-07-25.)
Headless browser across the APIs we track
| API | Published rate | Compared with a plain request | Checked |
|---|---|---|---|
| Decodo Web Scraping API | $0.75 per 1,000 requests | 1.5 × a plain request | Oct 6, 2026 |
| Firecrawl | 1 credit per page | 1 × a plain request | Oct 7, 2026 |
| HasData | 10 credits per request | 10 × a plain request | Oct 8, 2026 |
| Hyperbrowser | 1 credit per request | 1 × a plain request | Oct 5, 2026 |
| Olostep | 1 request per request | 1 × a plain request | Oct 9, 2026 |
| Scrape.do | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
| ScrapeLens | 20 credits per request | 1 × a plain request | Oct 9, 2026 |
| ScrapeOps | 10 credits per request | 10 × a plain request | Oct 9, 2026 |
| ScrapeOwl | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
| ScraperAPI | 10 credits per request | 10 × a plain request | Oct 9, 2026 |
| Scrapfly | 6 credits per request | 6 × a plain request | Oct 9, 2026 |
| ScrapingAnt | 10 credits per request | 10 × a plain request | Oct 9, 2026 |
| ScrapingBee | 5 credits per request | 5 × a plain request | Oct 7, 2026 |
| ScrapingBot | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
| ScrapingDog | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
| ScrapingFish | 1 request per request | 1 × a plain request | Oct 9, 2026 |
| SOAX Web Data API | 1 request, unit not published | not comparable to its plain rate | Sep 2, 2026 |
| WebScraping.AI | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
| ZenRows | 5 credits per request | 5 × a plain request | Oct 9, 2026 |
Several APIs render JavaScript at no extra charge and publish no separate rate; those are not listed here.
In more detail
The vendor examples below were read on the dates given in the text; the table above holds each vendor's current figure.
Headless-browser use changes a bill when the vendor meters time, traffic, concurrency, or browser operations rather than one returned page. A script that opens one session for ten pages can be billed differently from ten independent sessions. Idle time, rounding, proxy data, and failed interactions may still consume resources even if no usable record is returned. (Browserless unit consumption, Cloudflare Browser Rendering pricing, and Browserbase cost optimization, checked 2026-07-25.)
Browserless bills browser time from a shared Unit balance in started thirty-second increments. It adds separate Unit meters for datacenter or residential proxy data and for CAPTCHA attempts. A session lasting thirty-one seconds therefore consumes two browser-time increments before proxy or CAPTCHA use is counted. (Browserless unit consumption, checked 2026-07-25.)
Cloudflare Browser Rendering distinguishes Quick Actions from direct Puppeteer, Playwright, or Chrome DevTools Protocol sessions. Quick Actions use the browser-hour meter only. Direct sessions use browser hours plus a charge based on monthly-average concurrent browsers above the included amount, and monthly browser time is rounded to the nearest whole hour. (Cloudflare Browser Rendering pricing, checked 2026-07-25.)
Browserbase includes browser hours and concurrent browsers by plan, with residential proxy traffic metered separately. Each created session has a one-minute minimum, sessions can last up to six hours, and keep-alive sessions continue billing until explicitly released or timed out. (Browserbase pricing, Browserbase cost optimization, and Browserbase long sessions, checked 2026-07-25.)
ZenRows separates its Universal Scraper request modes from Scraping Browser usage. The Scraping Browser draws from the account balance using a browser-time meter and also incurs plan-specific bandwidth charges, while Universal Scraper uses per-request credit modes. Choosing the browser interface changes the billing basis. (ZenRows pricing rules, checked 2026-07-26.)
Before signing up, confirm the supported control protocol, when metering starts and stops, minimum and rounding rules, session timeout, concurrency, and proxy-data treatment. Estimate total active browser seconds and transferred bytes, not only URLs. Also check whether a simpler rendered-request endpoint can perform the job without maintaining browser state. (Browserless unit consumption, Cloudflare Browser Rendering pricing, and Browserbase cost optimization, checked 2026-07-25; ZenRows pricing rules, checked 2026-07-26.)
Sources
- docs.browserless.io/overview/unit-consumption
- developers.cloudflare.com/browser-run/
- developers.cloudflare.com/browser-run/pricing/
- www.browserbase.com/pricing
- docs.browserbase.com/optimizations/cost/cost-optimization
- docs.browserbase.com/platform/browser/long-sessions/timeouts
- docs.zenrows.com/first-steps/pricing