Releases: June 2025

Lawyers consistently deal with extensive documentation as a fundamental aspect of their practice. So the core part of casepal developments this month has also been concentrated on expanding the capabilities of our document processing infrastructure, enabling you to work with larger document sets and get better recognition of the scanned/handwritten documents.

‍

Constraints Mitigation

We have fundamentally restructured the Library's document processing architecture, moving from restrictive total page limits across all uploaded files to higher per-document capacity limits.

The comparison table below demonstrates the expansion in processing capabilities that now enables you to handle substantially larger and more complex legal documents within a single upload session.

‍

Processing Improvements

Previously, you might have occasionally experienced indexing failures in the Library that required manual deletion and re-uploading of documents to resolve processing issues.

Now, with our advanced processing pipeline, the vast majority of documents achieve successful indexing on the first attempt, eliminating the frustration of failed uploads and streamlining the document upload processes.

‍

Library Search Improvements

Previously, within the limited capacity of 200 pages per document, you might have encountered search retrieval limitations where casepal would primarily read through only the beginning portions of documents, sometimes failing to retrieve information located on later pages.

Now, library search functionality has been fundamentally refined, enabling casepal to easily and accurately retrieve information from all pages throughout the entire document, independent of length or complexity. The system now thoroughly analyses the full content of each document, ensuring that whether the relevant information appears on page 1 or page 1,500, you receive the necessary results with consistent accuracy and completeness.

‍

‍

Optical Character Recognition (OCR) improved

OCR is a technology used to recognize text in scanned documents and images. Even though we implemented this technology by design, we have been working continuously on the improvement of the recognition of scanned and handwritten texts from documents in different languages. Previously, OCR processing in the Library was notably limited, with documents starting to omit context after approximately 50 pages, making it impossible to process larger scanned legal documents effectively.

Now, we have completely overhauled our OCR processing capabilities so that scanned documents are processed with the same reliability and capacity as standard digital files. OCR documents now benefit from the same 1,500-page limit per document as regular files in the Library, eliminating the previous processing limitations and capacity restrictions.

Whether you are working with a 500-page Japanese scanned judgment, a 200-page Spanish contract with handwritten notes, or a 1,200-page scanned regulatory filing, casepal can now read, analyze, review, and accurately cite information from such documents without interruption or processing limitations.

‍

What's next?

And while we have done these important improvements to the processing of the documents, we have also been working continuously on the development of the long-anticipated proprietary tool of Assistant -  casepal Search.

The search enables  to conduct agentic searches across the web, with particular focus on primary legal frameworks, as well as other reputable secondary sources.

Web Search will appear live for users later this month, and will be followed by the release of the comprehensive instructions on usage and explanation of how casepal performs necessary legal research through publicly available legal sources.

‍