featherless-ai/Qwen3.5-4B-classifier

Hugging Face
VISIONPricing:Input $0.03 / Output $0Concurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32k Classification model Architecture:Transformer Warm

The featherless-ai/Qwen3.5-4B-classifier is a 4 billion parameter Qwen-based model, specifically designed for classification tasks using the Simple Jev classifier API on Featherless. It processes text, chat history, and images to provide structured decisions in JSON format, including choice, score, and yes/no (noul) classifications. This model excels at applications like intent detection, content moderation, and routing, offering a direct way to integrate AI-driven classification into applications.

Loading preview...

Qwen3.5-4B-classifier: Structured Decision-Making

The featherless-ai/Qwen3.5-4B-classifier is a 4 billion parameter Qwen-based model optimized for turning diverse contexts into structured decisions. It leverages the Simple Jev classifier API on Featherless to provide direct JSON outputs for classification tasks. This model is designed to streamline the integration of AI-powered classification into applications by allowing users to define questions and possible answers within each request.

Key Capabilities

  • Structured Classification: Converts text, chat conversations, or images into structured JSON outputs, including choices, scores, and yes/no propositions.
  • Flexible Question Types: Supports choice for selecting from named candidates, score for rating against an ordered rubric, and noul for yes/no propositions.
  • Multi-modal Context: Processes text, structured data, chat history, and images (vision capabilities).
  • Independent Question Evaluation: Allows multiple questions to be evaluated against the same context independently.

Good For

  • Support Routing: Automatically direct customer inquiries to the correct department.
  • Intent Detection: Identify user intentions from text or chat messages.
  • Content Moderation: Classify and filter content based on predefined criteria.
  • Relevance Scoring: Rate the relevance of information or items.
  • Agent Action Selection: Determine the next best action for an AI agent.
  • Visual Classification: Classify images based on specified criteria.