curl --request POST \
--url https://api.zerogpu.ai/v1/responses \
--header 'Content-Type: application/json' \
--header 'x-api-key: <api-key>' \
--data '
{
"input": "The customer says their payment failed and they want someone to call them.",
"model": "gliner2.5-decide-v1"
}
'{}{}gliner2.5-decide-v1
Model details for gliner2.5-decide-v1. Zero-shot decision and classification with caller-supplied labels.
curl --request POST \
--url https://api.zerogpu.ai/v1/responses \
--header 'Content-Type: application/json' \
--header 'x-api-key: <api-key>' \
--data '
{
"input": "The customer says their payment failed and they want someone to call them.",
"model": "gliner2.5-decide-v1"
}
'{}{}Fastino’s GLiNER2.5 Decide is a 486M-parameter zero-shot decision and classification model built for operational AI workflows. You supply the labels or decision options at request time to classify intent, route agents and tools, triage support requests, make moderation decisions, detect completion, and determine when a request needs human handoff. It handles single-label, multi-label, and several decision heads in one call, which suits high-volume routing and decision workloads where speed and structured outputs matter.References: Model docs • Terms • Privacy
Authorizations
Headers
Optional project identifier. Scopes the request to a specific project when provided.
Body
Model identifier (fixed for this playground). Use request examples to change use cases.
"gliner2.5-decide-v1""gliner2.5-decide-v1"
Multi-line text or document content to send to the model.
1 - 131072Use-case options for the model. Three use cases are supported:
ner— extract entities for the givenlabels.json— extract structured fields defined by aschema.classification— assign labels from candidate sets defined by aschema.
Hide child attributes
Hide child attributes
Which operation to run and which other fields apply: ner uses labels; json and classification use schema.
ner, json, classification Entity types to extract, used when usecase is ner. Each label becomes a key in the returned entities object.
[
"programming language",
"database",
"technology",
"protocol"
]
Minimum confidence score (0–1) an entity must reach to be returned (used with ner). Lower values surface more, lower-confidence matches.
0 <= x <= 1Extraction/classification definition, used when usecase is json or classification. For json, each group maps to field definitions in name::type::description form (e.g. "email::str::Email address"). For classification, each group maps to a list of candidate labels (e.g. ["positive", "negative", "neutral"]).
Hide child attributes
Hide child attributes
{
"contact": [
"name::str::Full name",
"email::str::Email address",
"phone::str::Phone number"
]
}
Response
Success
The response is of type object.