Adaptive Routing and Enhanced User Experience
This update introduces significant intelligence improvements to our automated routing engine, ensuring requests recover faster from temporary failures and avoid inefficient paths. We have also refined the authentication flow and resolved several edge-case bugs in request history and file management.
We've significantly increased the capacity of our data structuring pipeline. Complex page extractions now process much faster and remain stable even during sudden bursts of high-volume requests.
Our automated routing engine now adapts more quickly to changing site conditions and avoids becoming locked into outdated performance data. This ensures your requests consistently utilize the most effective paths for every domain.
Improvements
5Faster data extraction processing
We've significantly increased the capacity of our data structuring pipeline. Complex page extractions now process much faster and remain stable even during sudden bursts of high-volume requests.
General bug fixes and improvements
Plus 8 internal improvements for better reliability and performance.
Dynamic Routing Reliability
Our automated routing engine now adapts more quickly to changing site conditions and avoids becoming locked into outdated performance data. This ensures your requests consistently utilize the most effective paths for every domain.
Streamlined Signup Experience
The signup process now displays the email field by default, removing an unnecessary click for new users and creating a consistent experience across all login methods. You can now transition more quickly from account creation to your first request.
Simplified Capability Documentation
We have consolidated redundant entries in our anti-bot capability tables to provide a clearer overview of available features. This makes it easier for you to identify and utilize the specific tools required for your scraping needs.
Bug Fixes
7Reliable LLM data extraction
Fixed an internal configuration error that occasionally caused LLM-based data extraction requests to fail. These intelligent extraction requests now complete successfully and consistently return the expected schema.
Accurate playground request replays
Rerunning a previous scrape from the playground history now flawlessly preserves all original request parameters. This ensures that both the original HTTP method and any configured geo-targeting are correctly applied to the new run.
Consistent playground timing metrics
The response times shown in the playground's history panel now exactly match the accurate server-side processing metrics displayed in the result container. This provides a more accurate and unified view of your actual API latency.
Precise playground proxy toggling
Disabling the proxy toggle in the playground now completely removes geo-targeting from the underlying API payload. Previously, selecting a country would erroneously persist in the background even after the proxy feature was turned off.
Reliable support ticket attachments
Resolved an issue where file attachments uploaded while replying to an existing support ticket were occasionally dropped. Your diagnostic files and screenshots are now seamlessly attached to your support replies without silent failures.
Accurate Request History
Fixed an issue where POST requests restored from history would occasionally revert to GET after encountering a timeout or polling error. Your historical request data now accurately preserves HTTP methods across all states and error paths.
Improved File Upload Handling
We updated our file processing logic to correctly handle edge-case filenames that previously triggered server errors during the upload process. This ensures a smoother experience when managing attachments for support tickets and data exports.
Security
2Hardened support file uploads
Enhanced the strictness of backend validation and security for file uploads within the developer support portal. We have also improved the clarity of API error messages returned if an upload exceeds the allowed storage quota.
Accurate Resource Quota Tracking
Usage limits for file attachments are now calculated based on actual server-side measurements rather than client-provided metadata. This ensures fair resource allocation and prevents discrepancies in your account's quota reporting.
Plus 14 internal changes for stability and performance.