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

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF: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 Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF: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 Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Ollama
How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with Ollama:
```
ollama run hf.co/Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
```

Unsloth Studio

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF 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 Irfanuruchi/Qwen3-4B-Computer-Science-GGUF 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 Irfanuruchi/Qwen3-4B-Computer-Science-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

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

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF: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": "Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with Docker Model Runner:
```
docker model run hf.co/Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M
```

Lemonade

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-4B-Computer-Science-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Irfanuruchi/Qwen3-4B-Computer-Science-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Irfanuruchi/Qwen3-4B-Computer-Science-GGUF: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 Irfanuruchi/Qwen3-4B-Computer-Science-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

Qwen3-4B-Computer-Science-GGUF

GGUF release of Qwen3-4B-Computer-Science, a specialized language model fine-tuned for computer science, software engineering, Python programming, debugging, code generation, and technical reasoning.

This repository provides multiple GGUF quantizations for use with llama.cpp, LM Studio, Ollama, Jan, KoboldCpp, and other GGUF-compatible inference engines.

Model Overview

Base Model: Qwen/Qwen3-4B
Architecture: Qwen3
Format: GGUF
License: Apache-2.0
Language: English
Domain: Computer Science & Software Engineering

The model is designed to provide strong performance across software engineering tasks while remaining efficient enough to run locally on modern CPUs and GPUs.

Training Data

This model was fine-tuned using openly licensed datasets:

Dataset	License
HuggingFaceTB/smoltalk (smol-magpie-ultra)	Apache-2.0
agentica-org/DeepCoder-Preview-Dataset (primeintellect)	MIT

Training split:

60,989 samples

Evaluation split:

512 samples

Available Quantizations

File	Recommended Use
BF16.gguf	Highest quality, requires significant memory
Q8_0.gguf	Near-BF16 quality
Q6_K.gguf	Excellent quality/performance balance
Q5_K_M.gguf	Recommended for most users
Q4_K_M.gguf	Best memory efficiency

Recommended Quantization

For most systems:

Q5_K_M offers the best balance between:

Quality
Memory usage
Speed

If memory is limited, use Q4_K_M.

If maximum quality is desired, use Q8_0 or BF16.

Example (llama.cpp)

./llama-cli \
  -m Qwen3-4B-Computer-Science-Q5_K_M.gguf \
  -c 8192

Example (LM Studio)

Download one of the GGUF files.
Import the model into LM Studio.
Select the model.
Start chatting.

Example (Ollama)

Create a Modelfile:

FROM Qwen3-4B-Computer-Science-Q5_K_M.gguf

Then run:

ollama create qwen3-cs -f Modelfile
ollama run qwen3-cs

Intended Use

This model is intended for:

Software Engineering
Python Programming
Debugging
Code Review
Code Generation
Technical Question Answering
Algorithm Design
Computer Science Education

Limitations

This model is specialized for computer science tasks.

Performance outside software engineering domains may differ from the original base model.

As with all language models:

outputs may contain mistakes
generated code should be reviewed
security-critical code should always be validated

Integrity

SHA-256 hashes for every GGUF file are included in:

SHA256SUMS

Users are encouraged to verify downloaded files before use.

License

This repository is released under the Apache-2.0 License.

The fine-tuning datasets are compatible with commercial use:

Apache-2.0
MIT

Acknowledgements

Alibaba Qwen Team
Hugging Face
HuggingFaceTB
Agentica
llama.cpp contributors

Citation

If you use this model in your work, please cite this repository.

@misc{qwen3_4b_computer_science_gguf,
  title={Qwen3-4B-Computer-Science-GGUF},
  author={Irfanuruchi},
  year={2026},
  publisher={Hugging Face}
}

Downloads last month: 249

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for Irfanuruchi/Qwen3-4B-Computer-Science-GGUF

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Quantized

(283)

this model