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

Use Docker

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

LM Studio
Jan

vLLM

How to use Archsloth/Qwen3-4B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Archsloth/Qwen3-4B-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": "Archsloth/Qwen3-4B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

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

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

How to use Archsloth/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 Archsloth/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": "Archsloth/Qwen3-4B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

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

Lemonade

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

Pull the model

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

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Archsloth/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 Archsloth/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 "Archsloth/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

File size: 7,615 Bytes

<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 820 520" width="820" height="520" font-family="ui-sans-serif,-apple-system,'Segoe UI',Roboto,'Helvetica Neue',Arial,sans-serif" role="img" aria-label="File size against KL divergence, the ARCHsloth curve below the unsloth curve">
<rect width="820" height="520" fill="#0d0d0c"/>
<text x="34" y="40" fill="#f4f3ee" font-size="20" font-weight="700">Size &#215; quality &#183; Korean</text>
<text x="34" y="62" fill="#b6b4a8" font-size="12.5">Nine real files, no interpolation. Down and to the left is better &#8212; our whole curve sits under theirs.</text>
<rect x="34" y="80" width="13" height="13" rx="3" fill="#3987e5"/>
<text x="53" y="91" fill="#f4f3ee" font-size="12">ARCHsloth</text>
<rect x="150" y="80" width="13" height="13" rx="3" fill="#d95926"/>
<text x="169" y="91" fill="#f4f3ee" font-size="12">unsloth (their UD-*-XL files are <tspan font-weight="700">larger</tspan> than the same-named standard file)</text>
<line x1="84" y1="425.9" x2="772" y2="425.9" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="429.9" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.0010</text>
<line x1="84" y1="380.3" x2="772" y2="380.3" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="384.3" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.0020</text>
<line x1="84" y1="320.0" x2="772" y2="320.0" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="324.0" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.0050</text>
<line x1="84" y1="274.4" x2="772" y2="274.4" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="278.4" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.010</text>
<line x1="84" y1="228.8" x2="772" y2="228.8" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="232.8" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.020</text>
<line x1="84" y1="168.6" x2="772" y2="168.6" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="172.6" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.050</text>
<line x1="84" y1="123.0" x2="772" y2="123.0" stroke="#2a2a26" stroke-width="1"/>
<text x="74" y="127.0" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="end">0.100</text>
<text x="84.0" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">2.3</text>
<text x="219.2" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">2.9</text>
<text x="354.4" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">3.5</text>
<text x="489.6" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">4.1</text>
<text x="624.8" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">4.7</text>
<text x="760.0" y="468" fill="#807d72" font-size="11" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">5.3</text>
<text x="422.0" y="488" fill="#807d72" font-size="12" text-anchor="middle">file size (GB)</text>
<text transform="translate(19,278.0) rotate(-90)" fill="#807d72" font-size="12" text-anchor="middle">KL divergence (log scale)</text>
<line x1="125.2" y1="164.4" x2="125.2" y2="216.0" stroke="#2fae7d" stroke-width="1.6" stroke-dasharray="3 3"/>
<line x1="121.7" y1="164.4" x2="128.7" y2="164.4" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="121.7" y1="216.0" x2="128.7" y2="216.0" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="309.8" y1="298.2" x2="309.8" y2="333.4" stroke="#2fae7d" stroke-width="1.6" stroke-dasharray="3 3"/>
<line x1="306.3" y1="298.2" x2="313.3" y2="298.2" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="306.3" y1="333.4" x2="313.3" y2="333.4" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="532.1" y1="386.2" x2="532.1" y2="394.0" stroke="#2fae7d" stroke-width="1.6" stroke-dasharray="3 3"/>
<line x1="528.6" y1="386.2" x2="535.6" y2="386.2" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="528.6" y1="394.0" x2="535.6" y2="394.0" stroke="#2fae7d" stroke-width="1.6"/>
<line x1="105.2" y1="136.4" x2="131.2" y2="136.4" stroke="#6d6a61" stroke-width="2.5" stroke-dasharray="3 3"/>
<text x="140.2" y="130.4" fill="#807d72" font-size="11.5">stock <tspan font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace">llama-quantize</tspan></text>
<polyline points="118.2,164.4 129.4,173.2 302.8,298.2 383.1,335.9 525.1,386.2 702.3,410.4" fill="none" stroke="#d95926" stroke-width="2.5" stroke-linejoin="round"/>
<circle cx="118.2" cy="164.4" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="118.2" y="151.4" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">Q4_K_M</text>
<circle cx="129.4" cy="173.2" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="139.4" y="171.2" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="start">UD-Q4_K_XL</text>
<circle cx="302.8" cy="298.2" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="302.8" y="285.2" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">Q6_K</text>
<circle cx="383.1" cy="335.9" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="383.1" y="357.9" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">UD-Q6_K_XL</text>
<circle cx="525.1" cy="386.2" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="525.1" y="373.2" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">Q8_0</text>
<circle cx="702.3" cy="410.4" r="5.5" fill="#d95926" stroke="#0d0d0c" stroke-width="2"/>
<text x="702.3" y="432.4" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">UD-Q8_K_XL</text>
<polyline points="118.2,216.0 302.8,333.4 525.1,394.0" fill="none" stroke="#3987e5" stroke-width="2.5" stroke-linejoin="round"/>
<circle cx="118.2" cy="216.0" r="5.5" fill="#3987e5" stroke="#0d0d0c" stroke-width="2"/>
<text x="118.2" y="238.0" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">AS-Q4_K_M</text>
<circle cx="302.8" cy="333.4" r="5.5" fill="#3987e5" stroke="#0d0d0c" stroke-width="2"/>
<text x="302.8" y="355.4" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">AS-Q6_K</text>
<circle cx="525.1" cy="394.0" r="5.5" fill="#3987e5" stroke="#0d0d0c" stroke-width="2"/>
<text x="525.1" y="416.0" fill="#b6b4a8" font-size="11.5" font-family="ui-monospace,SFMono-Regular,Menlo,Consolas,monospace" text-anchor="middle">AS-Q8_0</text>
<text x="34" y="504" fill="#807d72" font-size="11.5">Green ticks join files of identical size. Six files came from unsloth's own repository; three were built here.</text>
</svg>