---
license: apache-2.0
language:
  - ko
  - en
tags:
  - korean
  - reasoning
  - instruction-tuning
  - fine-tuning
  - qwq
  - sft
---

# 🧠 QwQ-32B-Ko-Reasoning

> A large-scale Korean reasoning model fine-tuned from **Qwen/QwQ-32B**, designed to excel in logical and multi-hop reasoning tasks in Korean.

---

## 📌 Overview

**QwQ-32B-Ko-Reasoning** is a fine-tuned version of [Qwen/QwQ-32B](https://huggingface.co/Qwen/QwQ-32B), specifically optimized for **logical reasoning in Korean**. This model is part of a broader research initiative to explore:

- The **transition from multilingual reasoning LLMs** to **Korean-specialized reasoning models**
- The enhancement of **non-reasoning Korean language models** into **reasoning-capable variants**
- The development of open-access models that rival proprietary alternatives in complex reasoning tasks

This model was fine-tuned using a large-scale Korean-English instruction dataset containing diverse multi-hop questions, symbolic logic tasks, and human-crafted reasoning steps.

---

## 🧪 Benchmark Results

> - 📊 All benchmarks were measured using the **0-shot CoT (Chain-of-Thought)** method.
> - 📊 The **Score** represents either the **accuracy (%)** of correct answers or a rating on a **1-10 scale** from a judge model.
> - 📊 **LLM-as-a-judge** benchmarks were evaluated using **GPT-4o (2024-08-01-preview)**.

| **Benchmark**    | **Score**     |
|------------------|---------------|
| GPQA diamond     | 71.7          |
| GSM8K            | 74.6          |
| HAERAE           | 83.0          |
| KSM              | 83.1          |
| LogicKor         | 8.93          |
| Math500          | 85.3          |
| MT-Bench         | 8.36          |
| MT-Bench(Ko)     | 8.02          |

---

## 🧑‍💻 Usage

Install Transformers >= 4.50:

```bash
pip install -U transformers
```

Basic example:

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "DimensionSTP/QwQ-32B-Ko-Reasoning"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "서울과 부산 중 어디가 더 커?"
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)

model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=32768
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)
```

---

## 🧠 Base Model: Qwen/QwQ-32B

The base model, [Qwen/QwQ-32B](https://huggingface.co/Qwen/QwQ-32B), is a CoT LLM developed by the Alibaba Qwen team.
For more technical details, refer to the [Qwen2.5 Technical Report](https://arxiv.org/pdf/2412.15115).

---

## 🧱 Model Architecture

| Property         | Value                  |
|------------------|------------------------|
| Architecture     | Qwen2ForCausalLM       |
| Parameters       | 32B                    |
| Context Length   | 131,072 tokens         |
| Tokenizer        | QwenTokenizer (BPE)    |

---

## 📅 Release Date

**Mar 2025**  
This model was released in March 2025 as part of the **Ko-Reasoning Series**, which focuses on pushing the boundaries of open-source reasoning in Korean using modern LLMs.

---

## 📬 Contact

For questions, collaborations, or deployment inquiries, please contact:

- 🤖 Hugging Face: [https://huggingface.co/DimensionSTP](https://huggingface.co/DimensionSTP)
- ✉️ Email: [ddang8jh@gmail.com]

---

## 📦 Available Checkpoints

- ✅ `main`: Final stable version from the `last` branch
- ✅ All training artifacts available (tokenizer, config, model weights)