Skip to main content
POST
zlm-v1-multi-iab-classify: Responses
The instructions field is not needed for this model. Send only the text to classify. The result arrives as a JSON string in the response’s output text, so parse that string to read audience, content (iab_1_0 and iab_2_2), topics, keywords, and user_intent.
The multilingual variant of ZeroGPU’s IAB classifier returns the same full content-intelligence profile as the enriched edge model, for text that is not in English. One call gives you IAB categories across both the 1.0 and 2.2 taxonomies, audience and interest segments, topics, keywords, and user intent, every result scored for confidence. At 90M parameters it keeps the sub-millisecond profile the bidstream needs, so a multi-market publisher can run one classification pipeline instead of one per language. When your inventory spans languages and you still need clean IAB signals on every request, this is the model.
Supports batch processing through the Batch API. This model is a sibling of zlm-v2-iab-classify-edge-enriched; reach for that one when the text is English. References: TermsPrivacy

Authorizations

x-api-key
string
header
required

Headers

x-project-id
string

Optional project identifier. Scopes the request to a specific project when provided.

Body

application/json
model
string
default:zlm-v1-multi-iab-classify
required

Model identifier (fixed for this playground). Use request examples to change use cases.

Allowed value: "zlm-v1-multi-iab-classify"
Example:

"zlm-v1-multi-iab-classify"

input
string<textarea>
required

The text to classify, as a plain string, in any supported language. Inputs are truncated at 400 tokens.

Required string length: 1 - 131072

Response

Success

The response is of type object.