Named Entity Recognition
Hosted LLM named-entity extraction — pulls people, organisations, places, dates, and money references with much better recall than browser models on long-form text. Uses 1 credit per run.
About Named Entity Recognition
Named Entity Recognition pulls the people, organisations, places, dates, and money references out of text using a hosted LLM that recalls far more entities than lightweight browser models, especially across long documents. It's the right tool when you need a structured list of who, where, and how much from unstructured prose. Results come back as JSON and each run uses 1 credit.
- Category
- text
- Input
- Accepts: text/plain.
- Output
- Outputs: application/json.
- Cost
- Credit-metered
- Memory
- low
Common uses
- Extract every company and person named across a long article for a research dossier
- Pull dates and monetary amounts out of a contract for a quick at-a-glance summary
- Build a list of locations mentioned in a travel or logistics document
- Index meeting notes by the organisations and people discussed
- Surface the parties and figures in a financial report without reading it line by line
- Tag long-form content with its key entities for search or categorization
Frequently asked questions
What entity types does it find?
People, organisations, places, dates, and money references, returned as structured JSON.
Why use this over the free browser NER?
The hosted LLM has much better recall on long-form text, catching entities that browser-only models miss across longer passages.
What does it cost?
1 credit per run, since the extraction runs on a hosted model rather than in your browser.
Does my text leave my device?
Yes. The entities are extracted by a hosted model, so your text is sent there to be processed.
What input does it accept?
Plain text (text/plain). The output is JSON listing the entities it found.
Keywords
- ner
- entities
- extraction
- people
- orgs
- pro
- llm