Alelcv27/Qwen3-4B-INST-Math

TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kTool Calling:SupportedPublished:May 30, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

Alelcv27/Qwen3-4B-INST-Math is a 4 billion parameter instruction-tuned Qwen3 model developed by Alelcv27, fine-tuned for mathematical tasks. This model leverages Unsloth and Huggingface's TRL library for accelerated training, offering a 32768 token context length. It is optimized for efficient performance in mathematical reasoning and problem-solving applications.

Loading preview...

Model Overview

Alelcv27/Qwen3-4B-INST-Math is a 4 billion parameter instruction-tuned language model based on the Qwen3 architecture, developed by Alelcv27. This model was specifically fine-tuned for mathematical tasks, making it suitable for applications requiring numerical reasoning and problem-solving capabilities. It features a substantial context length of 32768 tokens, allowing it to process longer and more complex mathematical prompts.

Key Characteristics

  • Architecture: Qwen3-based, instruction-tuned.
  • Parameter Count: 4 billion parameters.
  • Context Length: Supports up to 32768 tokens.
  • Training Efficiency: Fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training.

Ideal Use Cases

  • Mathematical Reasoning: Excels in tasks requiring logical deduction and numerical computation.
  • Problem Solving: Suitable for applications involving complex mathematical problems.
  • Efficient Deployment: Its optimized training process suggests potential for efficient inference in mathematical contexts.