Robust Scraping and Standardized Limits
We have completely overhauled how SSL certificates and rate limits are handled across the platform. You can now scrape sites with invalid or expired certificates by default, utilize a new burst-allowance model for request limits, and benefit from standardized API error handling and faster error classification.
Our redesigned homepage now features an interactive live scrape demo. This lets you visualize real-time web scraping capabilities and instantly see data extraction results in action.
You can now scrape pages with expired, self-signed, or mismatched SSL certificates successfully without any extra configuration. These requests succeed automatically, drastically reducing failure rates on misconfigured websites.
Every rate limit across the platform now returns a standardized error envelope alongside exact rate-limit and retry headers. This allows your integrations to handle throttling elegantly and programmatically.
New Features
4Interactive Homepage and Scrape Demo
Our redesigned homepage now features an interactive live scrape demo. This lets you visualize real-time web scraping capabilities and instantly see data extraction results in action.
Ignore invalid SSL certificates by default
You can now scrape pages with expired, self-signed, or mismatched SSL certificates successfully without any extra configuration. These requests succeed automatically, drastically reducing failure rates on misconfigured websites.
Standardized 429 rate-limit responses
Every rate limit across the platform now returns a standardized error envelope alongside exact rate-limit and retry headers. This allows your integrations to handle throttling elegantly and programmatically.
Burst allowance for rate limits
We transitioned our rate limits to a token-bucket model that supports temporary traffic spikes. You can now send brief bursts of requests beyond your steady-state limit without being instantly throttled.
Improvements
7Faster Diagnostic API Responses
Our API diagnostics endpoint now uses enhanced caching and stricter type validation. This delivers faster, more reliable response data when monitoring your routing efficiency.
General bug fixes and improvements
Plus 8 internal improvements for better reliability and performance.
Detailed error causes for failures
Failed requests now include a structured underlying cause in the error payload for rendering and retrieval issues. This makes diagnosing failed attempts and tracking refunds straightforward.
Smarter error classification
We now distinguish true proxy issues from target website connection failures like DNS or SSL errors. This stops wasteful proxy rotations when a target site is down, delivering failure responses much faster.
Faster processing of complex HTML
Our parser has been optimized to handle nested or malformed markup in linear time. This eliminates performance bottlenecks and prevents timeouts when processing complex or poorly formatted pages.
Consistent request browser compatibility
Requests now employ a highly consistent browser identity and header configuration rather than randomizing profiles. This increases request predictability and significantly reduces the risk of detection.
Higher throughput for batch submissions
We removed restrictive edge-level IP rate limits from our batch and B2B job routers. This allows you to submit large volumes of scraping jobs from a single IP without encountering early throttling.
Bug Fixes
6Reliable Custom Action Validations
Creating custom actions via the API now strictly enforces input validation. This catches malformed execution parameters early, ensuring your scrape configurations run reliably.
Accurate Metric Regression Alerts
Failed metric probes no longer produce false-positive clean reports during regression monitoring. You will now receive accurate alerts if your baseline performance checks fail.
Automatic recovery for stalled jobs
Jobs assigned to worker queues are now automatically recovered if the assigned worker becomes unresponsive. This prevents requests from hanging indefinitely and ensures high overall task reliability.
Stable dashboard sessions
Temporary network drops or support widget disconnects will no longer log you out of your active dashboard session. Console sessions are now resilient to transient database and connection hiccups.
Accurate proxy error passthrough
API proxies now correctly forward upstream non-2xx status codes, detailed error bodies, and rate-limiting headers. This ensures your integration behaves identically when connecting via our proxies.
Improved dashboard error display
The developer console now robustly parses and displays all standard API error envelopes. This prevents page crashes and ensures you see clear, human-readable error messages instead of raw JSON.
Plus 25 internal changes for stability and performance.