# Changelog of OCR Text Extractor (`lergassy/ocr-text-extractor`) Actor

- **URL**: https://apify.com/lergassy/ocr-text-extractor/changelog.md
- **Full Actor documentation**: https://apify.com/lergassy/ocr-text-extractor.md

## Changelog

### 2026-10-18

- File hosts that refuse a browser-looking download from a datacenter (seen on w3.org: HTTP 403) no longer cost you the file: the download is retried with a plain bot User-Agent, then through a residential proxy. Missing files (404/410) are still reported straight away and not retried. The cost of each download step is saved in the `STATS` record.

### 2026-10-02

- Default memory now adapts to your input: small jobs use less, large batches get more.

### 2026-09-27

- Runs stop cleanly at your maximum cost instead of being aborted; nothing beyond your limit is written or charged.

### 2026-09-24

- Documentation improved.

### 2026-09-05

- Initial release: text from images and scanned PDFs with OCR in 19 languages, page by page, with an engine confidence score and optional word positions; no API key or external OCR service.