PRIVATE BY DEFAULT
Make sense of
your language.
Files stay on this device. No corpus upload, account or subscription.
0target corpus tokens0 files
Add your files or load the illustrative sample workspace to start.
Corpus manager
Session-only by default. Export a backup before clearing browser data.
Files, metadata & subcorpus selection
Tick target files to define a subcorpus. Enter research years and text categories for Trends / Text types. Years are not inferred from device timestamps.
TXT, DOCX, selectable PDF and CoNLL-U. Desktop: 250 MB / 50 million characters per file, 500 files per corpus, 100 million characters / 3 million indexed tokens across target + reference. Mobile/low-memory: 60 MB / 10 million characters, 250 files, 20 million characters / 1 million tokens. PDFs: up to 10,000 pages. Device memory affects practical capacity. Scans need OCR first.