featherless-ai/gemma-4-12B-it-classifier

Hugging Face
VISIONPricing:Input $0.24 / Output $0Concurrent Unit Cost:1Model Size:12BQuant:FP8Context Size:32k Classification model Architecture:Transformer Warm

The featherless-ai/gemma-4-12B-it-classifier is a 12 billion parameter Gemma-4 model, specifically designed for classification tasks using the Simple Jev classifier API. It excels at converting text, conversations, or images into structured decisions, providing JSON outputs for applications. This model is optimized for use cases like support routing, intent detection, content moderation, and relevance scoring, offering choice, score, and noul (yes/no) classification types.

Loading preview...

Model Overview

The featherless-ai/gemma-4-12B-it-classifier is a 12 billion parameter Gemma-4 model integrated with the Simple Jev classifier API on Featherless. This model is engineered to transform various forms of context—including text, conversations, and images—into structured, actionable decisions. It provides direct JSON outputs, making it suitable for automated application workflows.

Key Capabilities

  • Structured Decision Making: Converts input context into structured JSON results, including choices, scores, and yes/no propositions.
  • Flexible Question Types: Supports three primary classification types:
    • choice: Selects from named candidates with confidence and probabilities.
    • score: Provides an expected rubric index, confidence, and probabilities for ordered rubrics.
    • noul: Determines support for a yes/no proposition with a confidence value.
  • Multi-modal Context Support: Processes text, structured data, and chat history. Gemma classifier models also support image context.
  • Independent Question Evaluation: Allows multiple questions to be evaluated against the same shared context independently.

Good For

  • Support Routing: Automatically directs customer inquiries to appropriate departments.
  • Intent Detection: Identifies user intent from text or conversation.
  • Content Moderation: Classifies content for compliance or appropriateness.
  • Relevance Scoring: Rates the relevance of items or information.
  • Visual Classification: (For Gemma models) Classifies images based on defined criteria.
  • Agent Action Selection: Helps determine the next best action for an AI agent.