ArchiveStudio/Qwen3-4B

TEXT GENERATIONConcurrent Unit Cost:1Model Size:4BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jul 5, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

ArchiveStudio/Qwen3-4B is a 4 billion parameter causal language model from the Qwen series, developed by Qwen. It features a unique capability to seamlessly switch between a 'thinking mode' for complex reasoning tasks like math and code generation, and a 'non-thinking mode' for efficient general dialogue. With a native context length of 32,768 tokens, extendable to 131,072 tokens with YaRN, Qwen3-4B excels in enhanced reasoning, instruction-following, agent capabilities, and multilingual support across over 100 languages.

Loading preview...

Qwen3-4B Overview

Qwen3-4B is a 4 billion parameter causal language model, part of the latest Qwen series, developed by Qwen. It is distinguished by its innovative ability to dynamically switch between two operational modes:

Key Capabilities

  • Dual-Mode Operation: Uniquely supports seamless switching between a 'thinking mode' for complex logical reasoning, mathematics, and code generation, and a 'non-thinking mode' for efficient, general-purpose dialogue. This allows for optimized performance across diverse scenarios.
  • Enhanced Reasoning: Demonstrates significant improvements in reasoning capabilities, outperforming previous Qwen models in mathematics, code generation, and commonsense logical reasoning when in thinking mode.
  • Superior Human Preference Alignment: Excels in creative writing, role-playing, multi-turn dialogues, and instruction following, providing a more natural and engaging conversational experience.
  • Advanced Agent Capabilities: Integrates precisely with external tools in both thinking and non-thinking modes, achieving leading performance among open-source models in complex agent-based tasks.
  • Multilingual Support: Supports over 100 languages and dialects, offering strong capabilities for multilingual instruction following and translation.
  • Extended Context Length: Natively handles a context length of 32,768 tokens, which can be extended up to 131,072 tokens using the YaRN method for processing long texts.

When to Use This Model

  • Complex Problem Solving: Ideal for tasks requiring deep logical reasoning, such as mathematical problems or code generation, by leveraging its 'thinking mode'.
  • General Conversational AI: Suitable for efficient, general-purpose dialogue and instruction following in its 'non-thinking mode'.
  • Agentic Applications: Recommended for applications requiring tool integration and complex agent-based workflows due to its strong agent capabilities.
  • Multilingual Applications: A strong candidate for projects needing robust multilingual instruction following and translation across a wide array of languages.