cs-552-2026-baseline/group_model

TEXT GENERATIONConcurrency Cost:1Model Size:2BQuant:BF16Ctx Length:32kPublished:May 1, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

Qwen3-1.7B is a 1.7 billion parameter causal language model developed by Qwen, featuring a 32,768 token context length. This model uniquely supports seamless switching between a 'thinking mode' for complex logical reasoning, math, and coding, and a 'non-thinking mode' for efficient general-purpose dialogue. It demonstrates enhanced reasoning capabilities, superior human preference alignment for creative writing and role-playing, and strong agent capabilities with multilingual support for over 100 languages.

Loading preview...

Qwen3-1.7B Overview

Qwen3-1.7B is a 1.7 billion parameter causal language model from the Qwen series, designed with a 32,768 token context length. It introduces a novel feature allowing seamless switching between a 'thinking mode' for complex tasks and a 'non-thinking mode' for general dialogue, optimizing performance across diverse scenarios.

Key Capabilities

  • Dynamic Thinking Modes: Uniquely switches between a reasoning-focused mode (for math, code, logical reasoning) and an efficient general-purpose mode.
  • Enhanced Reasoning: Shows significant improvements in mathematics, code generation, and commonsense logical reasoning compared to previous Qwen models.
  • Human Preference Alignment: Excels in creative writing, role-playing, multi-turn dialogues, and instruction following, providing a natural conversational experience.
  • Agentic Expertise: Integrates precisely with external tools in both thinking and non-thinking modes, achieving leading performance in complex agent-based tasks among open-source models.
  • Multilingual Support: Supports over 100 languages and dialects with strong multilingual instruction following and translation capabilities.

Best Practices for Usage

  • Sampling Parameters: Different Temperature, TopP, TopK, and MinP settings are recommended for thinking vs. non-thinking modes to prevent issues like endless repetitions. Greedy decoding is discouraged for thinking mode.
  • Output Length: Recommend 32,768 tokens for most queries, extending to 38,912 for highly complex problems like math and programming competitions.
  • Standardized Output: Use specific prompts for math problems (e.g., "Please reason step by step, and put your final answer within \boxed{}") and multiple-choice questions (e.g., JSON structure for answers) to standardize model outputs during benchmarking.