Dario213/Qwen3-4B-medical-reasoning

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Mar 11, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

Dario213/Qwen3-4B-medical-reasoning is a 4 billion parameter Qwen3 model developed by Dario213, fine-tuned for medical reasoning tasks. This model leverages LoRA adapters and was trained using Unsloth and Huggingface's TRL library for accelerated fine-tuning. It is specifically optimized for complex medical reasoning, utilizing the FreedomIntelligence/medical-o1-reasoning-SFT dataset. The model has a context length of 32768 tokens, making it suitable for processing extensive medical texts.

Loading preview...

Model Overview

Dario213/Qwen3-4B-medical-reasoning is a 4 billion parameter language model developed by Dario213, specifically fine-tuned for medical reasoning. It is based on the Qwen3 architecture and was trained using Unsloth and Huggingface's TRL library, which enabled 2x faster training.

Key Capabilities

  • Medical Complex Reasoning: The model is fine-tuned on the FreedomIntelligence/medical-o1-reasoning-SFT dataset, making it proficient in handling complex medical reasoning tasks.
  • Efficient Training: Utilizes LoRA adapters on all modules with a rank of 8, contributing to efficient fine-tuning.
  • Optimized Training Parameters: Trained with specific SFTConfig arguments including warmup_steps=5, learning_rate=2e-4, optim="adamw_8bit", weight_decay=0.001, and lr_scheduler_type="linear".
  • Extended Context: Features a context length of 32768 tokens, allowing for the processing of longer medical documents and complex case studies.

Good For

  • Applications requiring medical question answering and diagnostic support.
  • Research and development in AI-driven medical reasoning.
  • Tasks involving the analysis and synthesis of medical literature and patient data.