CoPE-B: The COntent Policy Evaluator Model
Model Overview
CoPE is a family of special purpose content policy evaluation models designed to perform accurate content labeling based on developer-customizable criteria. Built by trust & safety veterans, CoPE represents a significant advancement in content classification technology, offering steerable and precise policy interpretation capabilities while maintaining a small and efficient model size.
CoPE-B-12b-it is the latest model in the CoPE family and supports multimodal labeling (both text and images), a much longer context window of 128k tokens, and instructiong tuning (i.e., chat interface). CoPE-B-12B-it-gemma utilizes Google's gemma-3-12b-it as its backbone model. The current version is a release candidate and may not be the final model Zentropi releases for CoPE-B.
Note: This model is currently only available to paid subscribers of the Zentropi platform. If you are not a subscriber and would like access to the open version of CoPE, please download CoPE-A-9B.