Instructions to use cortexso/qwen-qwq with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cortexso/qwen-qwq with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="cortexso/qwen-qwq",
	filename="qwq-32b-q2_k.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use cortexso/qwen-qwq with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/qwen-qwq:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/qwen-qwq:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cortexso/qwen-qwq:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf cortexso/qwen-qwq: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 cortexso/qwen-qwq:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf cortexso/qwen-qwq: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 cortexso/qwen-qwq:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cortexso/qwen-qwq:Q4_K_M

Use Docker

docker model run hf.co/cortexso/qwen-qwq:Q4_K_M

LM Studio
Jan

vLLM

How to use cortexso/qwen-qwq with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/cortexso/qwen-qwq:Q4_K_M

Ollama
How to use cortexso/qwen-qwq with Ollama:
```
ollama run hf.co/cortexso/qwen-qwq:Q4_K_M
```

Unsloth Studio new

How to use cortexso/qwen-qwq 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 cortexso/qwen-qwq 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 cortexso/qwen-qwq to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for cortexso/qwen-qwq to start chatting

Pi new

How to use cortexso/qwen-qwq with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf cortexso/qwen-qwq: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": "qwen-qwq"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use cortexso/qwen-qwq with Docker Model Runner:
```
docker model run hf.co/cortexso/qwen-qwq:Q4_K_M
```

Lemonade

How to use cortexso/qwen-qwq with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cortexso/qwen-qwq:Q4_K_M

Run and chat with the model

lemonade run user.qwen-qwq-Q4_K_M

List all available models

lemonade list

Overview

QwQ is the reasoning model of the Qwen series. Unlike conventional instruction-tuned models, QwQ is designed to think and reason, achieving significantly enhanced performance in downstream tasks, especially challenging problem-solving scenarios.

QwQ-32B is the medium-sized reasoning model in the QwQ family, capable of competitive performance against state-of-the-art reasoning models, such as DeepSeek-R1 and o1-mini. It is optimized for tasks requiring logical deduction, multi-step reasoning, and advanced comprehension.

The model is well-suited for AI research, automated theorem proving, advanced dialogue systems, and high-level decision-making applications.

Variants

No	Variant	Cortex CLI command
1	QwQ-32B	`cortex run qwen-qwq:32b`

Use it with Jan (UI)

Install Jan using Quickstart
Use in Jan model Hub:
```
cortexso/qwen-qwq
```

Use it with Cortex (CLI)

Install Cortex using Quickstart
Run the model with command:
```
cortex run qwen-qwq
```

Credits

Author: Qwen Team
Converter: Homebrew
Original License: License
Paper: Introducing QwQ-32B: The Medium-Sized Reasoning Model

Downloads last month: 95

GGUF

Model size

33B params

Architecture

qwen2

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit