Named entity recognition, running on your own machine

bert-base-NER labels people, organisations, locations and miscellaneous entities with character offsets you can copy straight into a pipeline. The Piiranha model on the same page goes further and detects personal information — emails, phone numbers, account numbers, addresses — which is exactly the kind of text you should not be sending to a hosted API in the first place. Both run in the browser, so redaction happens before anything crosses the network.

named entity recognition onlinener tool freebert-base-ner demopii detection browserentity extraction toolredact pii locallytoken classification online

Frequently asked questions

Can I use this to redact documents before sending them elsewhere?

That is the intended use of the Piiranha model, and it is a sensible workflow because the detection runs locally. Always review the output — no NER model catches everything, especially unusual identifier formats.

Which entity types are supported?

bert-base-NER emits PER, ORG, LOC and MISC in CoNLL-2003 style. Piiranha covers a wider set of personal-data categories including emails, phone numbers and account identifiers.

Models available for this task

BERT Base NER — Xenova/bert-base-NER, 110 MB, 108M params
mBERT NER (10 languages) — Xenova/bert-base-multilingual-cased-ner-hrl, 182 MB, 177M params
DistilBERT NER (10 languages) — Xenova/distilbert-base-multilingual-cased-ner-hrl, 139 MB, 135M params
Piiranha v1 (PII) — onnx-community/piiranha-v1-detect-personal-information-ONNX, 334 MB, 280M params
BERT Small PII — onnx-community/bert-small-pii-detection-ONNX, 30 MB
DistilBERT NER — onnx-community/distilbert-NER-ONNX, 67 MB, 65M params

Explore the rest of the playground

Text & language models

LLM chat — Chat with a local LLM running on your own GPU
Text embeddings — Generate text embeddings locally, in the browser
Translation — Translate text offline with a local neural MT model
Summarization — Summarise long text with a model running on your machine
Sentiment analysis — Sentiment analysis that runs in your browser
Zero-shot text classification — Classify text into labels you make up on the spot
Question answering — Ask questions about a passage, get the exact answer span
Fill-mask — Watch a masked language model guess the missing word

Guides & comparisons

Model catalog — The model catalog: what actually runs in a browser
WebGPU vs WASM — WebGPU vs WebAssembly for running models in a browser
Browser AI vs cloud API — Browser inference vs a hosted API: an honest comparison
Transformers.js vs WebLLM — Transformers.js vs WebLLM
Whisper model sizes — Choosing a Whisper size for browser transcription
fp32 vs q8 vs q4 — Picking a quantization level for browser inference
Why no image generation — Why this playground has no text-to-image generation
FAQ — Frequently asked questions about browser-based AI
About — About this project

This playground needs JavaScript and WebGPU. Models are downloaded to your browser and executed on your own GPU, so there is nothing to show without a script engine. Enable JavaScript, or read the FAQ for what the site does.