Can Infinite Graph import a scanned PDF?
Not reliably. The current parser does not provide OCR, and scanned or encrypted PDFs can be reported as import errors. A text-readable PDF is the safer input for this workflow.
DOCUMENTS / CONNECTED READING
Turn a PDF into a knowledge graph.
THE SHORT ANSWER
A PDF-to-knowledge-graph workflow turns a document's structure into connected, explorable nodes instead of leaving its relationships implicit inside one file. Infinite Graph accepts PDF imports, keeps source and page context with the imported material, and lets you review the resulting graph while AI assists with summaries and relationship suggestions that you decide whether to keep.
Open Infinite GraphTHE PROBLEM
A PDF preserves information, but not the path between ideas.
A paper, report, or book can contain definitions, evidence, references, and recurring concepts without making those connections easy to revisit. A folder of PDFs preserves the files; it does not give you a visual workspace for comparing concepts across documents or tracing an idea back to its source.
A REVIEWABLE WORKFLOW
From source document to reviewed graph
The useful distinction is not automatic extraction versus no extraction. It is whether the extracted structure stays inspectable and editable.
- 01 / SOURCE
Import the PDF
Choose a PDF from the supported document-import workflow and keep the original file as the source context.
- 02 / STRUCTURE
Preserve sections and pages
Parsing and normalization produce structured units with document metadata, page locations, and diagnostics where the source needs review.
- 03 / EXPLORE
Search and inspect connections
Move from a document view to a visual graph, semantic search, cluster summaries, and bounded graph context when those features are available.
- 04 / DECIDE
Keep the relationships you trust
AI can propose summaries or connections, but the workspace stays yours: review, edit, arrange, or leave a suggestion unused.
PRODUCT EVIDENCE
What the product can substantiate
This page describes the current product boundary rather than promising a perfect semantic conversion of every PDF.
- The document model includes PDF, Markdown, HTML, DOCX, EPUB, and plain-text formats.
- Imported structure can carry a source document name, revision identity, page location, content hash, and parser diagnostics.
- Semantic retrieval, graph traversal, and cluster summaries support exploration after import; they are not a replacement for reading the source.
- PDF parsing does not provide OCR for scanned documents, and encrypted or scanned PDFs can be reported as import errors.
IN PRACTICE
Example: compare a small literature set
A researcher imports several papers about the same question, follows recurring concepts across their sections, and uses semantic search to return to the relevant part of a paper. Suggested links become useful when they make a relationship easier to inspect—not when they are accepted without review.
Illustrative workflow, not a measured product result.
A USEFUL DISTINCTION
PDF summary, mind map, or knowledge graph?
A summary compresses prose into a shorter explanation. A mind map is useful when one topic should branch into a clear hierarchy. A knowledge graph is a better fit when you need to move between concepts, documents, and sources without forcing everything into one tree.
- Use a summary when the main task is fast recall.
- Use a mind map when one central topic and a small hierarchy are enough.
- Use a knowledge graph when relationships across sections or documents are part of the question.
QUESTIONS WORTH ANSWERING
Start with the boundary, not the buzzword.
Does the PDF become a final, authoritative knowledge graph automatically?
No. Imported structure and source context provide a starting point. AI-generated summaries or relationship suggestions remain proposals until you review and adopt or edit them.
What other document formats are supported?
The document model includes plain text, Markdown, HTML, DOCX, EPUB, and PDF. The exact result can vary by format and source quality.
Can I change the graph after importing a PDF?
Yes. Infinite Graph is a visual workspace: you can search, move through relationships, arrange the graph, and decide which suggested changes belong in your own knowledge space.
WHEN YOU ARE READY
Bring the next question into the graph.
Open Infinite Graph in the browser and keep the decisions about your knowledge with you.
Open Infinite Graph