APEX FLOW WEB SCRAPER / REFERENCE LIBRARY
HTTP scraping vs. browser rendering
HTTP collection is the lighter starting point for server-rendered pages. A browser is useful when JavaScript creates the required content, but it costs more memory and time.
Choose based on the missing field
Look at the received HTML before enabling a browser. If the title and price are already present, rendering adds work without improving those fields. If a page contains only an application shell, JavaScript may be necessary.
Apex rendering modes
never uses HTTP extraction; always renders; auto attempts the HTTP path and may render when content is insufficient. Rendering uses an installed Brave browser in an isolated context, not your personal signed-in session. HTTP failure and access challenges are reported rather than converted into success.
Rendering has boundaries
The current engine blocks downloads, WebSockets, service workers, image/media/font loads and methods other than GET or HEAD. Resource limits and these restrictions can prevent some applications from working. Login workflows, CAPTCHA solving and authenticated account scraping are outside this release.
Compare fairly
Measure end-to-end job time, extraction accuracy, browser starts, transferred bytes and memory. A browser benchmark and an HTTP-only benchmark answer different questions. Do not call one faster without naming the workload.
← All reference articles · Current capabilities and limits →