About Split2AI
Split2AI is a browser-based preprocessing tool for preparing oversized PDFs for AI services such as ChatGPT, Claude and Gemini.
What problem does it solve?
Large manuals, technical documents, papers and scanned books can run into file-size or page-count limits. Published service limits also do not guarantee that every PDF below a limit will process reliably. Split2AI aims to reduce both the mechanical work of splitting a document and the uncertainty around how much preprocessing is sensible.
Privacy principle
The core PDF splitting work happens in the browser. Split2AI is not designed to upload PDF contents to its server in order to split them. Product measurement is also designed not to include PDF contents, page text or filenames.
How evidence is separated
The AI PDF limits page keeps published service limits separate and records their source context.
The real-device benchmark publishes device, size, page count and verified output steps without turning one successful test into a universal guarantee.
The AI PDF fit checker labels Split2AI's practical size guidance separately from official limits.
The methodology explains how specifications, observations and operational guidance are kept distinct.
What exists today
The current core is browser-local PDF splitting. OCR, chapter-aware splitting and other preprocessing ideas are future candidates; they are not presented as available features before they exist. The PDF preprocessing guide explains when splitting, compression or OCR addresses a particular failure mode.
Policies
See the privacy policy and terms for the public operating rules.
Open Split2AI