PDF, Word, CSV and images
Reads native PDF text, Word documents, text and CSV files, with OCR for scanned pages.
DocLens
Ask your documents, get answers with sources.
Upload a report, contract or spreadsheet and ask questions in plain language. DocLens finds the relevant passages and answers from them, citing the page each answer came from.
It runs on OjasYukti's own server with a self-hosted AI model, so documents are not sent to an outside AI provider. Your uploads are private to your browser and are deleted automatically after 7 days.
Private to your browser · Deleted after 7 days · Self-hosted AI
What it does
Built for reading, not guessing.
Answers come from the document you uploaded. When the answer is not in it, DocLens says so.
Hybrid search
Combines meaning-based search with exact keyword matching, so both concepts and specific names or clause numbers are found.
Cited answers
Each answer points back to the source document and page so it can be checked.
Exact table answers
Tables are extracted, then filtered, counted or totalled with real calculations instead of estimates.
Follow-up questions
Ask a follow-up such as “what about last year?” and the conversation context is kept.
Private by default
Documents and chats are visible only in the browser that uploaded them, and are deleted automatically after 7 days.
Grounded answers
The model answers from your document, not from memory.
Before answering, DocLens retrieves the most relevant passages from the uploaded file and gives only those to the AI model. Numbers in tables are computed rather than predicted.
Check important answers against the cited page, especially for legal, financial or medical documents.
How it works
01 Upload a PDF, Word file or spreadsheet02 Wait while it is read and indexed03 Ask a question in plain language04 Check the answer against its cited page
Have a document to dig into?
DocLens is available at doclens.ojasyukti.tech.