Okada0311/dpo-qwen-cot-merged

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

Okada0311/dpo-qwen-cot-merged is a 4 billion parameter language model, fine-tuned from Qwen/Qwen3-4B-Instruct-2507 using Direct Preference Optimization (DPO). This model is specifically optimized to enhance reasoning capabilities through Chain-of-Thought (CoT) and improve the quality of structured responses. It is designed for applications requiring robust logical processing and coherent, well-structured text generation.

Loading preview...

Model Overview

Okada0311/dpo-qwen-cot-merged is a 4 billion parameter language model derived from Qwen/Qwen3-4B-Instruct-2507. It has been fine-tuned using Direct Preference Optimization (DPO) with the Unsloth library, integrating the full 16-bit weights directly into the base model, eliminating the need for adapter loading.

Key Capabilities

  • Enhanced Reasoning: Optimized through DPO to improve Chain-of-Thought (CoT) reasoning, enabling more logical and step-by-step problem-solving.
  • Improved Response Quality: Focuses on generating higher quality, more structured, and aligned outputs based on a preference dataset.
  • Direct Use: As a fully merged model, it can be used directly with the transformers library without additional configuration for LoRA adapters.

Training Details

The model underwent 1 epoch of DPO training with a learning rate of 5e-06 and a beta value of 0.1. It utilized a maximum sequence length of 512 tokens and was trained on the u-10bei/dpo-dataset-qwen-cot dataset. The license for this model follows the MIT License, with users also required to comply with the original base model's license terms.

Good For

  • Applications requiring strong reasoning and logical inference.
  • Generating structured and coherent text responses.
  • Tasks where response quality and alignment with preferred outputs are critical.