Accelerated Scrapes and Precise Billing
This release significantly accelerates scraping speeds by streamlining our queue handoffs and skipping unnecessary processing steps for raw data requests. We have also enhanced billing and status accuracy by automatically detecting blocked pages to prevent false charges and preserving actual upstream HTTP status codes in structured results. Additionally, our automated extraction engine is now more resilient with self-healing capabilities around underlying AI models.
Structured scraping requests now complete up to one second faster. By replacing polling intervals with instant queue handoffs, your extracted data is delivered immediately after the page is scraped.
Scrape requests that only require raw HTML responses now completely bypass the structuring queue. This optimization removes unnecessary processing to deliver your raw content significantly faster.
You will not be billed for scrapes that encounter block pages, challenges, or system errors. Our detection systems now identify these occurrences to ensure false successes are refunded immediately.
Improvements
5Instant structured job handoffs
Structured scraping requests now complete up to one second faster. By replacing polling intervals with instant queue handoffs, your extracted data is delivered immediately after the page is scraped.
Bypassed structuring for raw scrapes
Scrape requests that only require raw HTML responses now completely bypass the structuring queue. This optimization removes unnecessary processing to deliver your raw content significantly faster.
Reduced peak latency tail
We capped individual retry timeouts for advanced browser-based requests. This eliminates the long latency tail, ensuring slow requests either failover rapidly or resolve within a predictable time limit.
Resilient schema extraction engine
Our automatic data extraction engine now automatically self-heals when encountering offline AI model providers. This protects your parsing pipelines from external service disruptions.
General bug fixes and improvements
Plus 15 internal improvements for better reliability and performance.
Bug Fixes
3Billing protection for blocked pages
You will not be billed for scrapes that encounter block pages, challenges, or system errors. Our detection systems now identify these occurrences to ensure false successes are refunded immediately.
Preserved upstream status codes
Structured scrape responses now return the actual HTTP status code from the targeted upstream website rather than defaulting to 200. This enables more precise error handling in your application.
Accurate scrape outcome tracking
Scrapes that return successful responses from non-200 pages, like soft-404s, are no longer falsely marked as failed in your usage analytics. This ensures your dashboard metrics are accurate.
Plus 15 internal changes for stability and performance.