Instructions to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ISTA-DASLab/Qwen3-4B-GGUF-GSQ",
	filename="qwen3-4b-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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
# Run inference directly in the terminal:
llama-cli -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
# Run inference directly in the terminal:
llama-cli -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
# Run inference directly in the terminal:
./llama-cli -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Use Docker

docker model run hf.co/ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

LM Studio
Jan

vLLM

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ISTA-DASLab/Qwen3-4B-GGUF-GSQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ISTA-DASLab/Qwen3-4B-GGUF-GSQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Ollama
How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with Ollama:
```
ollama run hf.co/ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
```

Unsloth Studio new

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ 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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ 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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ISTA-DASLab/Qwen3-4B-GGUF-GSQ to start chatting

Pi new

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

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": "ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

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 ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Run Hermes

hermes

Docker Model Runner
How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with Docker Model Runner:
```
docker model run hf.co/ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K
```

Lemonade

How to use ISTA-DASLab/Qwen3-4B-GGUF-GSQ with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Run and chat with the model

lemonade run user.Qwen3-4B-GGUF-GSQ-Q2_K

List all available models

lemonade list

Qwen3-4B — GGUF K-Quant refined with GSQ

GGUF K-Quant checkpoints of Qwen/Qwen3-4B in which the discrete grid assignments have been refined with GSQ (Gumbel-Softmax Quantization), starting from a public GGUF initialization and projected back into the same K-Quant format. The optimized files run unchanged on llama.cpp / Ollama.

Paper: GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling (arXiv:2604.18556)
Paper page on HF: https://huggingface.co/papers/2604.18556
Code: https://github.com/IST-DASLab/GSQ
Collection: https://huggingface.co/collections/ISTA-DASLab/gsq

Quantization details

Base model: Qwen/Qwen3-4B
Format: GGUF K-Quant (drop-in replacement for standard K-Quant files)
Pipeline: GGUF K-Quant init → GSQ Gumbel-Softmax refinement → re-pack into K-Quant
Runtime: llama.cpp, ollama, LM Studio, anything that consumes GGUF

Storage layout

These files are bit-for-bit standard GGUF K-Quant. GSQ only changes the values of the quantized weights (it relearns the discrete grid assignments inside each K-Quant block), not the block structure, scales, or super-block layout. As a result:

The file size matches the corresponding upstream K-Quant for the same quant tier.
Any llama.cpp / ollama build that loads regular Qwen3-4B-Q2_K.gguf loads this file with zero changes.
The Hugging Face UI reports GGUF block types (e.g. Q2_K, Q4_K, Q6_K) rather than per-tensor dtypes — those refer to the on-disk K-Quant encoding, not the precision of any optimizer state.

Usage with llama.cpp

huggingface-cli download ISTA-DASLab/Qwen3-4B-GGUF-GSQ \
  Qwen3-4B-Q2_K.gguf --local-dir .

./llama-cli -m Qwen3-4B-Q2_K.gguf -p "Hello"

Usage with Ollama

ollama run hf.co/ISTA-DASLab/Qwen3-4B-GGUF-GSQ:Q2_K

Citation

@article{gsq2026,
  title  = {GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling},
  author = {Dadgarnia, Alireza and Tabesh, Soroush and Nikdan, Mahdi and Helcig, Michael and Kurti{\'c}, Eldar and Kleinegger, Max and Alistarh, Dan},
  journal= {arXiv preprint arXiv:2604.18556},
  year   = {2026},
  url    = {https://arxiv.org/abs/2604.18556}
}