text Faster on WebGPU

Named Entity Recognition

Identify people, organizations, and locations in text — runs on your device.

First run downloads ~105 MB. The model is cached after the first use, then runs offline. Manage downloads on the settings page.
Loading…

About Named Entity Recognition

Named Entity Recognition scans text and pulls out the people, organizations, and locations mentioned in it, all processed on your device. Reach for it when you need to know who and what a document talks about — building an index, tagging notes, or extracting structured facts from prose. It returns the entities as JSON, and the text never leaves your browser.

Category
text
Input
Accepts: text/plain.
Output
Outputs: application/json.
Cost
Free, runs in your browser
Memory
medium
Install group
nlp-standard
Privacy: Named Entity Recognition runs entirely on your device. Files you provide never leave your browser — no uploads, no server, no tracking. The page works offline once loaded.

Common uses

  • Pull every person and company name out of meeting minutes to build an attendee or stakeholder list
  • Tag a batch of news articles by the organizations and places they mention
  • Extract location names from travel notes to plot or organize later
  • Build a quick index of named parties from a contract or report draft
  • Identify the key players mentioned across a set of emails
  • Generate metadata tags for a document library based on the entities inside each file

Frequently asked questions

Which entity types does it find?

People, organizations, and locations. The result is returned as JSON listing each entity it identifies.

Is my text uploaded?

No. The NER model runs locally in your browser, so the document you analyze stays on your device.

How is this different from keyword extraction?

Keyword extraction surfaces topic words and noun phrases. NER specifically identifies named entities — actual people, companies, and places — rather than general subject terms.

What input does it accept?

Plain text. Paste your prose or feed in the output of a text-extraction tool.

Does it catch dates and money amounts too?

This free version focuses on people, organizations, and locations. For dates, money, and broader entity types with higher recall, use the hosted NER Pro tool.

How accurate is it on long documents?

It works well on clear prose, but recall drops on very long or unusual text. The hosted Pro version has notably better recall on long-form content.

Keywords

  • ner
  • entities
  • person
  • organization
  • location
  • nlp
  • extract

Try next