Extract named entities from text using GLiNER PII model

post

https://integrate.api.nvidia.com/v1/chat/completions

Extract named entities and PII (Personally Identifiable Information) from input text.
The model supports custom entity labels and returns entities with positions, scores,
and tagged text output. The API is OpenAI-compatible, wrapping entity extraction
results in a chat completion format.

Body Params

model

string

Defaults to nvidia/gliner-pii

The GLiNER model to use for entity extraction

messages

array of objects

required

length ≥ 1

A list of messages with the text to analyze. Only the last user message
content will be used for entity extraction.

Messages*

role

string

enum

required

The role of the message author

Allowed:

content

string

required

For user messages: the text to analyze for entity extraction.
For assistant messages: JSON string containing entity extraction results
with fields: total_entities, entities (array of {text, label,
start, end, score}), and tagged_text.

labels

array of strings

Entity types to detect. None uses the default PII labels
(55 categories including email, phone_number,
ssn, first_name, last_name, address, etc.).

Labels

threshold

number

0 to 1

Defaults to 0.5

Confidence threshold for entity detection (0.0 to 1.0).
Lower values detect more entities but may include false positives.

chunk_length

integer

1 to 2048

Defaults to 384

Context window size for processing. Longer texts are
automatically split into chunks with overlap for complete coverage.
Must be greater than overlap.

overlap

integer

0 to 512

Defaults to 128

Token overlap between chunks to prevent entity clipping. Must be less than chunk_length.

flat_ner

boolean

Defaults to false

When True, prevents overlapping entity spans. When False, may return
nested entities (e.g., both "John Doe" as a name and "John" as first_name).

Responses

Language

Credentials

Bearer

Response

Click Try It! to start a request and see the response here! Or choose an example:

application/json

200Entity extraction completed successfully

202Result is pending. Client should poll using the requestId.

422Validation failed, provided entity could not be processed.

500The invocation ended with an error.