Instructions to use BryanADA/Qwen2.5-3B-cot-zh-tw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="BryanADA/Qwen2.5-3B-cot-zh-tw")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("BryanADA/Qwen2.5-3B-cot-zh-tw", dtype="auto")

llama-cpp-python

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="BryanADA/Qwen2.5-3B-cot-zh-tw",
	filename="Qwen2.5-3B-cot-zh-tw.Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Use Docker

docker model run hf.co/BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

LM Studio
Jan

vLLM

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BryanADA/Qwen2.5-3B-cot-zh-tw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BryanADA/Qwen2.5-3B-cot-zh-tw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

SGLang

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "BryanADA/Qwen2.5-3B-cot-zh-tw" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BryanADA/Qwen2.5-3B-cot-zh-tw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "BryanADA/Qwen2.5-3B-cot-zh-tw" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BryanADA/Qwen2.5-3B-cot-zh-tw",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Ollama:
```
ollama run hf.co/BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
```

Unsloth Studio

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for BryanADA/Qwen2.5-3B-cot-zh-tw to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for BryanADA/Qwen2.5-3B-cot-zh-tw to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for BryanADA/Qwen2.5-3B-cot-zh-tw to start chatting

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Docker Model Runner:
```
docker model run hf.co/BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M
```

Lemonade

How to use BryanADA/Qwen2.5-3B-cot-zh-tw with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull BryanADA/Qwen2.5-3B-cot-zh-tw:Q4_K_M

Run and chat with the model

lemonade run user.Qwen2.5-3B-cot-zh-tw-Q4_K_M

List all available models

lemonade list

Qwen-2.5-3B-cot-zh-tw (GRPO)

模型簡介 | Model Overview

本模型基於 Qwen-2.5-3B-Instruct，專為繁體中文數學/邏輯推理場景設計，不是單純仿製長鏈推理，而是經由創新 RLHF 訓練流程，讓模型自發產生類似「aha moment」的推理能力。

訓練流程為：

以少量高品質多步推理 SFT 數據冷啟動模型
接著採用 GRPO 策略與自設獎勵函數，在「問題—答案」數據集下強化模型「自發推理」與「步驟合理性」，不依賴 PPO 或大規模 SFT
推理長度經本地測試優化，步驟數落在最佳甜蜜點，適合部署於一般 GPU/邊緣裝置

This model is based on Qwen-2.5-3B-Instruct and optimized for step-by-step reasoning in Traditional Chinese math and logic tasks. Instead of standard CoT SFT or PPO, we use a minimal SFT “cold start” with high-quality reasoning samples, then apply GRPO with a custom reward function to let the model discover its own “aha moments” and multi-step reasoning chains—all while controlling output length for efficient local deployment.

訓練動機 | Motivation

這個專案純粹出於個人興趣與繁體中文圈的實際需求而開發，旨在讓本地端能有更好用的推理模型。我希望能為繁體中文圈貢獻可以本地部署的推理模型，讓模型在繁中語境下，也能自發產生多步驟、具備頓悟感（aha moment）的解題過程。整體訓練過程強調實驗精神與實用導向：只用少量優質 SFT 冷啟動，再透過 GRPO 與自定獎勵函數，讓模型自己學會思考與拆解問題，追求可實際落地的推理效果。

This project was developed purely out of personal interest and the practical needs of the Traditional Chinese-speaking community, with the goal of creating a more effective locally deployable reasoning model. I hope to contribute a model that enables multi-step, “aha moment” reasoning in the Traditional Chinese context—one that can be run efficiently on local hardware. The entire training process emphasizes both experimentation and real-world usability: starting with a small amount of high-quality SFT for cold start, then leveraging GRPO and custom reward functions to encourage the model to learn reasoning and problem decomposition on its own, with a focus on truly applicable step-by-step solutions.

模型特性 | Key Features

Aha Moment 自發推理：非模板複製，而是訓練模型「自己發現推理步驟」

步驟最佳化：推理長度經本地實測，控制在「解題效率」與「可讀性」甜蜜點

繁體中文強化：涵蓋台灣、港澳常用語境，數學、邏輯均表現穩定

適用本地端部署：硬體需求親民，適合一般 4GB GPU，步驟控制精簡，不會無限發散

Self-generated Reasoning: Not pattern imitation—model “discovers” solution steps

Optimized Step Count: Output length tuned for real-world efficiency and readability

Traditional Chinese Enhanced: Robust for Taiwan/HK/Macau context and math/logic QA

Local Deployment Friendly: Runs on standard consumer GPUs, with step count sweet spot

訓練細節 | Training Details

基礎模型 / Base：Qwen2.5-3B-Instruct

流程 / Pipeline：少量高品質多步推理 SFT（冷啟動）→ GRPO（自訂獎勵函數，自發推理強化）

數據來源 / Data：DoggiAI/GSM8K_zh_tw + 自建繁體推理 Q&A (From Grok API Distill)

RLHF 核心：獎勵重點放在答案正確率、推理步驟合理性與精簡性，不靠人類標記每步驟

硬體 / Hardware：L4 GPU, 訓練 24 小時

框架 / Framework：Transformers, PEFT, bitsandbytes, Unsloth

使用建議 | Usage Tips

推薦應用：數學解題、邏輯題、逐步問答適合使用類似：「請自行分步推理，說明每一步的原因。」等提示語可將 output 步驟數量根據需求進行控制（建議 2~6 步為最佳）

快速上手 | Quickstart

print("Hello, world!")
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("BryanADA/qwen-2.5-3b-cot-zh-tw")
tokenizer = AutoTokenizer.from_pretrained("BryanADA/qwen-2.5-3b-cot-zh-tw")

prompt = "小明有 3 顆蘋果，又拿到 2 顆，一共幾顆？請分步說明推理過程。"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(
    **inputs,
    max_new_tokens=512,
    do_sample=True,
    temperature=0.7,
    top_p=0.95,
    pad_token_id=tokenizer.eos_token_id
)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

參考資料 | References

Qwen 官方

Deepseek R1 論文

DoggiAI/GSM8K_zh_tw

RLHF/GRPO 相關文獻

License

本模型採用 Apache-2.0 授權，允許用於研究、學術及商業用途。請遵循授權條款保留原作者版權及免責聲明。

This model is licensed under Apache-2.0, allowing use for research, academic, and commercial purposes. Please comply with the license terms and retain the original copyright and disclaimers.

Downloads last month: 8

GGUF

Model size

3B params

Architecture

qwen2

Hardware compatibility

4-bit

Model tree for BryanADA/Qwen2.5-3B-cot-zh-tw

Base model

Qwen/Qwen2.5-3B

Finetuned

Qwen/Qwen2.5-3B-Instruct

Quantized

(223)

this model

BryanADA
/

Qwen2.5-3B-cot-zh-tw