Classify Text

jina.classify

ActionAI & modelsconnection required

Calls the Jina AI API with the connection's credentials.

Classify inputs zero-shot against candidate labels, or with a classifier previously trained on this account.

Ask about Jina AI All Jina AI nodes Provider API docs ↗

Connection

Required: Yes — the node will not save without one
Auth type: Bearer token
Providers: Jina AI
Secret shape: the API key, or {"api_key": "…"}
Config it needs: base_url, default_embedding_model, default_reranker_model
Resolved by: config.connection_id, else node.connection_id

Configuration

4 fields, 1 required. A dotted path is a nested object and name[] an array of objects — the editor renders the same tree.

Field	Type	Required	Default	Constraints	Example
input Items to classify: up to 512 strings or {"text": "…"} objects.	array	Yes	—	—	["the invoice is overdue"]
labels Candidate labels for zero-shot classification — up to 512 of them. Leave empty when using a trained classifier_id.	array	No	—	array of string	["billing","support"]
model Embedding model backing the zero-shot classifier, for example jina-embeddings-v3. Falls back to the connection's default.	string	No	—	—	jina-embeddings-v3
classifier_id Use a trained few-shot classifier instead of labels.	string	No	—	—	—

At least one of labels, classifier_id must be set.

Example configuration

Every value below is the contract’s own example or default — nothing invented.

{
  "input": [
    "the invoice is overdue"
  ],
  "labels": [
    "billing",
    "support"
  ],
  "model": "jina-embeddings-v3"
}

Output

One shape.

result

Field	Type	Example
predictions One result per input item, in input order.	array	—
predictions[].index The item's position in the input.	number	—
predictions[].prediction The winning label.	any	—
predictions[].score Confidence in the winning label.	any	—
predictions[].predictions Every candidate label with its score.	array	—
count How many items were classified.	number	—
usage Token usage Jina billed for this call.	object	—

{
  "count": 0,
  "predictions": [
    {
      "index": 0,
      "prediction": null,
      "predictions": [],
      "score": null
    }
  ],
  "usage": {}
}

Related Jina AI nodes

See Classify Text run in a workflow.

In a technical demo we add the Jina AI connection, drop the node on the canvas and let Workflow Health pre-flight it before the first real run.

Book a technical demo Talk to sales