Instructions to use QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QubaxAI/Qwen3-4B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/QubaxAI/Qwen3-4B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use QubaxAI/Qwen3-4B-GGUF with Ollama:
```
ollama run hf.co/QubaxAI/Qwen3-4B-GGUF:Q4_K_M
```
Unsloth Desktop

How to use QubaxAI/Qwen3-4B-GGUF with Pi:

Start the llama.cpp server

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

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "QubaxAI/Qwen3-4B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

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

Lemonade

How to use QubaxAI/Qwen3-4B-GGUF with Lemonade:

Pull the model

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

Run and chat with the model

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

List all available models

lemonade list

Hermes Agent

How to use QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-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 QubaxAI/Qwen3-4B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use QubaxAI/Qwen3-4B-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf QubaxAI/Qwen3-4B-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 "QubaxAI/Qwen3-4B-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"

Qwen3-4B GGUF — ready-to-run quantized builds

Pre-quantized GGUF builds of Qwen/Qwen3-4B for use with llama.cpp, Ollama, LM Studio, and any GGUF-compatible runtime.

We maintain these so you can run a strong 4B model locally in minutes — and when you need 340+ frontier models instead, our API at qubax.ai serves them at some of the lowest prices on the market (pay with crypto, no KYC, no credit card).

Available quantizations

File	Quant	Size	Quality
`Qwen3-4B-Q4_K_M.gguf`	Q4_K_M	~2.4 GB	Best balance — recommended default
`Qwen3-4B-Q8_0.gguf`	Q8_0	~4.7 GB	Near-lossless

Quick start (llama.cpp)

# Download
huggingface-cli download QubaxAI/Qwen3-4B-GGUF Qwen3-4B-Q4_K_M.gguf --local-dir .

# Run (llama.cpp)
llama-server -m Qwen3-4B-Q4_K_M.gguf --port 8080

Or with Ollama:

ollama run hf.co/QubaxAI/Qwen3-4B-GGUF:Q4_K_M

Why Qwen3-4B?

Hybrid thinking / non-thinking modes in one model
Strong reasoning + multilingual performance for its size
Runs on ~4 GB VRAM at Q4 — fine on a laptop

Who are we?

Qubax AI is an AI model marketplace: 340+ models, one OpenAI-compatible API, priced up to 99% below standard retail. Crypto payments, no subscription, no KYC.

🌐 qubax.ai

License

Apache 2.0 (inherited from the base model). Model by the Qwen team — this repo only redistributes quantized weights.

Downloads last month: 69

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for QubaxAI/Qwen3-4B-GGUF

Base model

Qwen/Qwen3-4B-Base

Finetuned

Qwen/Qwen3-4B

Quantized

(321)

this model