Chattso-GPT/dpo-qwen-cot-merged

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Feb 2, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

Chattso-GPT/dpo-qwen-cot-merged is a 4 billion parameter Qwen3-based instruction-tuned causal language model, fine-tuned using Direct Preference Optimization (DPO). It is specifically optimized to improve reasoning capabilities through Chain-of-Thought (CoT) and generate high-quality structured responses. This model is ideal for applications requiring enhanced logical reasoning and structured output generation.

Loading preview...

Overview

This model, Chattso-GPT/dpo-qwen-cot-merged, is a 4 billion parameter language model based on the Qwen/Qwen3-4B-Instruct-2507 architecture. It has undergone Direct Preference Optimization (DPO) using the Unsloth library, resulting in a full-merged 16-bit weight model that requires no adapter loading.

Key Capabilities

  • Enhanced Reasoning: Optimized through DPO to improve Chain-of-Thought (CoT) reasoning, making it more effective for complex logical tasks.
  • Structured Responses: Fine-tuned to align its outputs with preferred formats, leading to higher quality structured responses.
  • Efficient Deployment: Provided as a fully merged model, simplifying deployment with standard transformers library usage.

Training Details

The model was trained for 1 epoch with a learning rate of 1e-07 and a beta of 0.1, using a maximum sequence length of 1024. The training utilized the u-10bei/dpo-dataset-qwen-cot dataset, which focuses on preference alignment for reasoning and structured output. The model's license is MIT, with compliance required for the original base model's terms.