Compare two PDFs. Inspect the sources.
Find recurring and newly mentioned topics with original page citations. No PDF uploads. Not an automated eligibility decision.
Your two PDF files
Choose two text-based PDFs. The comparison runs locally without registration. Keyword coverage is strongest for German-language tenders; other languages may not be detected completely. Original source passages stay untranslated. Explore the public TED demo first ↗
Restores both reports, your evidence inventory and all manual notes together. Original PDF file bytes are not included.
Privacy: your documents stay on your device. Details
The evidence list and PDF report are processed in your browser. Files are only created when you explicitly download JSON or CSV. No account, AwardCove cloud storage or external AI. Anonymous action categories may be counted without document content or personal identifiers. Keep separate client workspaces.
Add evidence or reopen saved reports
These controls are not required to compare two PDFs. Use them for a manually managed evidence inventory or to resume a saved review.
Optional · Local evidence inventory
List evidence documents and any expiration date. Certificate PDFs are not uploaded.
Import replaces the current inventory and closes a previously loaded PDF report for safety. Never mix reports from different clients.
Evidence entries: 0
Optional · Compare a saved report
Export the JSON result using AwardCove PDF checker (German) and reopen it here. The PDF itself is not transferred.
No report imported yet.
Other comparison methods and the public demo
No personal files needed: two official, publicly accessible TED award notices are compared for example text mentions (TED 280759-2026 and 297207-2026). Not complete procurement packs, not a real client, and no proven reuse of the same legal evidence. Loading the sample replaces the current local session after confirmation.
Load an earlier saved PDF report (JSON). It stays in your browser and is compared with the later report.
Responsibility: no legally binding go/no-go decision and no completeness guarantee. The research workflow does not track personally identifiable use. Data transparency.