Match the words
Typo-tolerant search across names, titles, authors, tags and the full text of every document, with the matching words highlighted in the result.
Try “expense reimbursment”
Ask a question in plain English and get an answer grounded in your documents, with citations that open the exact passage behind each statement.
One search box over every file your team is allowed to see. Look for the exact words, describe what you mean, or do both at once.
Typo-tolerant search across names, titles, authors, tags and the full text of every document, with the matching words highlighted in the result.
Try “expense reimbursment”
Describe what you are looking for in your own words. Results are ranked by how close they are in meaning, even when they share no words with your question.
Try “getting paid back for a client trip”
Words and meaning are fused into a single score, so a result that matches both rises to the top. The mode to reach for when you are not sure how it was written.
Try “deadline to claim taxi expenses”
Two views of every search
The Documents view answers the first question: one result per file, with its summary, category and tags.
The Passages view answers the second: the exact section that matches, with its heading and page. Open passage takes you to that place in the document, not to page one.
Narrow any search by workspace, category, tags, file type or language. Every value shows how many results sit behind it, so you always know what a click will leave you with.
The words, the mode and the filters all live in the address bar. Bookmark a search you run often, or send it to a colleague and they see exactly what you saw.
Semantic and hybrid search use the AI provider you configure. Keyword search works on its own.
Every document travels the same pipeline on its own, and you can follow each step on screen. Nothing has to be tagged, converted or organized by hand first.
A file is uploaded, registered or found in a watched folder.
Text, pages and details such as author and language are read from the file.
AI writes a summary and keywords and suggests a category for the document.
The document and each of its passages are indexed in LipiCore Engine.
Ready for search and for the AI Assistant, in the workspace it belongs to.
Drag and drop one file or many. Each uploads with its own progress, and you can name, tag and categorize it on the way in.
Files already on your server or a mounted share are indexed where they are. Nothing is copied or moved.
Point a workspace at a shared folder. New and changed files are picked up on a schedule, and files that are removed drop out of search.
Also built in: duplicate detection by content, so the same file is never indexed twice in a workspace.
Documents live in workspaces, and each person or group is granted the workspaces they need. A search or an AI answer only ever draws on what that person is allowed to see.
Keep HR policies, contracts, finance records or project files in workspaces of their own. Switch a workspace off and it leaves search at once, without deleting a single file.
Decide who can read each workspace, who can upload and edit, and who can use the AI Assistant. Set it per person, or once on a group that everyone in the team belongs to.
The same accounts work across LipiCore Insights and LipiCore Documents. Sign in once, move between the two, and connect your own identity provider for single sign-on.
LipiCore Documents is a white-label document search and AI Assistant. It reads the files your organization already has, indexes them in LipiCore Engine, and lets your team search them by words or by meaning and ask questions answered from the documents themselves.
PDF, Word, Excel, PowerPoint, plain text, HTML and Markdown, along with image, audio and video files. Scanned PDFs and images are read through OCR when it is enabled for your deployment.
No. LipiCore Documents runs inside your own infrastructure. You can upload files to it, register files that already sit on your server, or point it at a shared folder that it checks on a schedule. Files registered in place are indexed where they are and never copied.
Yes. Without an AI provider, every document is still read and indexed, and keyword search, filters and passages work as usual. Search by meaning, automatic summaries and the AI Assistant switch on when you configure the AI provider of your choice.
Yes. Documents are kept in workspaces, and access is granted per user or per group. A person only searches, reads and gets AI answers from the workspaces they have been given.
Yes. LipiCore Documents is white label and uses your logo, colors and product name. It shares its branding and theme with LipiCore Insights, so the two look like one product to your team.
Yes. LipiCore Documents runs on top of LipiCore Engine, which provides the index that search and the AI Assistant work from. LipiCore Insights is optional and can run beside it on the same sign-in.
Get in touch with our team to see how LipiCore Documents can make the files your business runs on searchable and answerable.