Compare · Scrapling
Webfold vs Scrapling
On a page that needs JavaScript, Webfold answered in 3.3 ms. Scrapling's Chromium took 907.6 ms. That is about 275 times faster at the median. On a plain fetch Scrapling was faster, 9.3 ms against 12.0.
Scrapling is a Python library with a stealth fetcher that drives Chromium. Webfold is a Rust program with its own CDP server. We ran three jobs on local fixtures. Each ran 15 times. None failed.
Webfold on its own
How fast is Webfold alone?
Each job is measured on its own. The performance page states the figures once. We link them here.
- Run a page's scripts and drive it like a browser: 8 ms to start, 70.7 ms a scripted task. 106.6 tasks a second in 30 MB.
- Turn a web page into clean Markdown: 4.6 µs a page, in process. About 217,000 pages a second per core, conversion only.
Now the comparison
Webfold against Scrapling
| Aspect | Webfold | Scrapling |
|---|---|---|
| Page that needs JavaScript, median | 3.3 ms (own interpreter over CDP) | 908 ms (StealthyFetcher on Chromium) |
| Static parse, 2,000-row table, median | 7.2 ms (a process per document) | 8.4 ms (warm Python process) |
| Plain HTTP fetch then extract, median | 12.0 ms | 9.3 ms (curl_cffi, Chrome impersonation) |
| Cost of starting the library | A single binary. | About 120 ms to import the parser alone. |
| Browser TLS impersonation | Not established. | Yes, through curl_cffi. |
| Language | Rust, single binary, CLI and MCP. | Python library. |
Where is each tool ahead?
Where Scrapling is ahead
- Plain fetches. Scrapling was faster here (9.3 ms against 12.0 ms median).
- Browser TLS fingerprint impersonation. A mature stealth fetcher.
- A large Python ecosystem. A flexible selector API.
Where Webfold is ahead
- Pages that run JavaScript: about 275 times faster at the median. The engine is not a browser. That gap is the browser, not Scrapling's code.
- No interpreter start-up per call. A one-shot CLI call does not pay Python's import cost.
- One binary. No Python environment to manage.
How do I move from Scrapling to Webfold?
| If you use | In Webfold |
|---|---|
| Fetcher.get + Selector | webfold scrape <url>, or POST /v1/web/scrape |
| StealthyFetcher for script-built pages | webfold headless serve and a CDP client |
What is the evidence?
Each figure states its dataset. It states its limits.
FAQ
Frequently asked questions
Does Webfold bypass bot protection like Scrapling's stealth mode?
Not reliably. Stealth is still an open item for us, and we do not claim TLS fingerprint parity. Use Scrapling for hard targets and Webfold for the pages that do not need it.
Why is the JavaScript lane so lopsided?
Scrapling launches Chromium for that lane and Webfold runs the page's script in its own engine, which starts in milliseconds. That engine passes the full independent agent-browser benchmark; see the compatibility page for what it does not yet run.
Try it on your own pages.
Run the command-line tool. Join the waitlist for the hosted API.