Zero-Shot NER (GLiNER) Node

Z

AI/ML/ONNX/NLP

Extract entities for any labels you name at runtime, with no fixed label set and no retraining. Load a GLiNER ONNX export (e.g. https://huggingface.co/onnx-community/gliner_small-v2.1, gliner_multi-v2.1, gliner_medium_news-v2.1, gliner_multi_pii-v1, NuNER_Zero) plus the tokenizer.json from the same repository. For models with a fixed label set, use the Named Entity Recognition node instead.

onnx_glineronnx

Inputs9

Outputs4

Security exposureUnrated

Packageonnx

Ratings

Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.

No score metadata has been set for this node yet.

Input Pins

9

Input

Execution

exec_in

Initiate Execution

Model

Struct

model

ONNX GLiNER Model Session

SchemaNodeOnnxSessionNodeOnnxSessionNodeOnnxSession1 fields

session_refstringrequired

Cache ID for Session

Schema enforced

Tokenizer

Struct

tokenizer

HuggingFace tokenizer.json from the same model repository

SchemaFlowPathFlowPathFlowPath3 fields

pathstringrequired

store_refstringrequired

cache_store_refstring | null

Schema enforced

Text

String

text

Input text to analyze for named entities

Labels

String Array

labels

Entity types to look for, in plain language (e.g. person, company, medication, invoice number)

Threshold

Float

threshold

Minimum confidence for a span to be reported (0.0-1.0)

Default 0.5

Range 0 to 1

Max Span Width

Integer

max_width

Longest entity in words. Must match the model's max_width from gliner_config.json (12 for most GLiNER models, 1 for NuNER Zero)

Default 12

Multi Label

Boolean

multi_label

Report every label that clears the threshold for a span instead of only the best one

Default false

Merge Adjacent

Boolean

merge_adjacent

Join neighbouring same-label entities separated only by whitespace. Required for token-level models such as NuNER Zero, which score one word at a time

Default true

Output Pins

4

Output

Execution

exec_out

Done

Result

Struct

result

Full zero-shot result with entities and the labels that were requested

SchemaGlinerResultGlinerResultGlinerResult4 fields

entitiesArray<GlinerEntity>required

Accepted entities, ordered by position

itemsGlinerEntityarray item

A span GLiNER assigned one of the runtime labels

textstringrequired

The entity text, sliced from the input

labelstringrequired

The runtime label this span matched

start_charinteger:uintrequired

Character start position in the original text

format uintmin 0

end_charinteger:uintrequired

Character end position in the original text (exclusive)

format uintmin 0

start_wordinteger:uintrequired

Index of the first word in the span

format uintmin 0

+2 more fields

textstringrequired

Original input text

labelsArray<string>required

Labels the model was asked about

itemsstringarray item

word_countinteger:uintrequired

Words the text was split into

format uintmin 0

Entities

Struct Array

entities

Extracted entities as array

SchemaGlinerEntityGlinerEntityGlinerEntity7 fields

textstringrequired

The entity text, sliced from the input

labelstringrequired

The runtime label this span matched

start_charinteger:uintrequired

Character start position in the original text

format uintmin 0

end_charinteger:uintrequired

Character end position in the original text (exclusive)

format uintmin 0

start_wordinteger:uintrequired

Index of the first word in the span

format uintmin 0

end_wordinteger:uintrequired

Index of the last word in the span (inclusive)

format uintmin 0

scorenumber:floatrequired

Sigmoid score for this span/label pair

format float

Count

Integer

entity_count

Number of entities found

Node Info

Internal name

onnx_gliner

Category

AI/ML/ONNX/NLP

Version

1