adrianxu9778/Qwen2.5-3B-Instruct

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:3.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 28, 2026License:otherArchitecture:Transformer Featherless Exclusive Cold

adrianxu9778/Qwen2.5-3B-Instruct is a 3.09 billion parameter instruction-tuned causal language model from the Qwen2.5 series, developed by Qwen. This model features a transformer architecture with a 32,768 token context length and is significantly improved in coding, mathematics, and instruction following. It excels at generating long texts, understanding structured data like JSON, and offers robust multilingual support across 29 languages.

Loading preview...

Qwen2.5-3B-Instruct: An Enhanced Language Model

Qwen2.5-3B-Instruct is an instruction-tuned causal language model with 3.09 billion parameters, part of the latest Qwen2.5 series. Developed by Qwen, this model builds upon its predecessors with substantial improvements across several key areas, making it a versatile tool for various applications.

Key Capabilities and Improvements

  • Enhanced Knowledge & Reasoning: Significant advancements in general knowledge, coding, and mathematics, leveraging specialized expert models.
  • Superior Instruction Following: Greatly improved ability to follow instructions, generate long texts (up to 8K tokens), and understand/generate structured data, including JSON.
  • Robustness to System Prompts: More resilient to diverse system prompts, enhancing role-play and chatbot condition-setting.
  • Extended Context Length: Supports a full context length of 32,768 tokens, with generation capabilities up to 8,192 tokens.
  • Multilingual Support: Comprehensive support for over 29 languages, including major global languages like Chinese, English, French, Spanish, German, and Japanese.
  • Architectural Foundation: Utilizes a transformer architecture incorporating RoPE, SwiGLU, RMSNorm, Attention QKV bias, and tied word embeddings.

Ideal Use Cases

This model is particularly well-suited for scenarios requiring:

  • Code Generation and Mathematical Problem Solving: Due to its specialized training in these domains.
  • Complex Instruction Following: For applications needing precise adherence to user commands.
  • Long-form Content Generation: Capable of producing extensive and coherent text outputs.
  • Structured Data Processing: Efficiently handles and generates structured formats like JSON.
  • Multilingual Applications: Effective across a broad spectrum of languages for global deployment.

Finetunes
1 models