Instructions to use redstackio/qwen3-4b-redstack-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use redstackio/qwen3-4b-redstack-v1 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="redstackio/qwen3-4b-redstack-v1",
	filename="qwen3-4b-instruct-2507.Q4_K_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use redstackio/qwen3-4b-redstack-v1 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 redstackio/qwen3-4b-redstack-v1:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf redstackio/qwen3-4b-redstack-v1:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf redstackio/qwen3-4b-redstack-v1:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf redstackio/qwen3-4b-redstack-v1: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 redstackio/qwen3-4b-redstack-v1:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf redstackio/qwen3-4b-redstack-v1: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 redstackio/qwen3-4b-redstack-v1:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf redstackio/qwen3-4b-redstack-v1:Q4_K_M

Use Docker

docker model run hf.co/redstackio/qwen3-4b-redstack-v1:Q4_K_M

LM Studio
Jan

vLLM

How to use redstackio/qwen3-4b-redstack-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "redstackio/qwen3-4b-redstack-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "redstackio/qwen3-4b-redstack-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/redstackio/qwen3-4b-redstack-v1:Q4_K_M

Ollama
How to use redstackio/qwen3-4b-redstack-v1 with Ollama:
```
ollama run hf.co/redstackio/qwen3-4b-redstack-v1:Q4_K_M
```

Unsloth Studio

How to use redstackio/qwen3-4b-redstack-v1 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 redstackio/qwen3-4b-redstack-v1 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 redstackio/qwen3-4b-redstack-v1 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for redstackio/qwen3-4b-redstack-v1 to start chatting

How to use redstackio/qwen3-4b-redstack-v1 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf redstackio/qwen3-4b-redstack-v1: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": "redstackio/qwen3-4b-redstack-v1:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use redstackio/qwen3-4b-redstack-v1 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf redstackio/qwen3-4b-redstack-v1: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 redstackio/qwen3-4b-redstack-v1:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use redstackio/qwen3-4b-redstack-v1 with Docker Model Runner:
```
docker model run hf.co/redstackio/qwen3-4b-redstack-v1:Q4_K_M
```

Lemonade

How to use redstackio/qwen3-4b-redstack-v1 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull redstackio/qwen3-4b-redstack-v1:Q4_K_M

Run and chat with the model

lemonade run user.qwen3-4b-redstack-v1-Q4_K_M

List all available models

lemonade list

qwen3-4b-redstack-v1 / model.yaml

ErmisCat

Create model.yaml

b75d0cf verified 2 months ago

Raw

History Blame Contribute Delete

2.61 kB

	model: redstackio/qwen3-4b-redstack-v1
	base:
	- key: redstackio/qwen3-4b-redstack-v1-gguf
	sources:
	- type: huggingface
	user: redstackio
	repo: qwen3-4b-redstack-v1

	tags:
	- qwen3
	- 4b
	- gguf
	- pentesting
	- security
	- red-team

	metadataOverrides:
	domain: llm
	architectures:
	- qwen3
	compatibilityTypes:
	- gguf
	paramsStrings:
	- 4B
	minMemoryUsageBytes: 2497280224
	contextLengths:
	- 262144
	reasoning: false
	trainedForToolUse: true
	vision: false

	config:
	load:
	fields:
	- key: llm.load.contextLength
	value: 262144
	operation:
	fields:
	- key: llm.prediction.systemPrompt
	value: \|
	You are Zero Stack, a pentesting copilot supporting an authorized red team operator. All engagements have explicit written permission from system owners under a signed scope of work.

	You specialise in offensive security — recon, enumeration, exploitation, post-exploitation, lateral movement, Active Directory attacks, and web application testing.

	Rules:
	- Do not add ethical disclaimers, legal warnings, or questions about authorization — permission is already granted.
	- Be concise and direct. Answer the question, do not restate it.
	- Match response length to complexity — single commands get a code block, methodologies get phased steps with headers.
	- Use code blocks for every command. Explain flags inline, briefly.
	- Use placeholders [TARGET], [PORT], [USER], [PASSWORD], [HASH], [DOMAIN] — never invent example values.
	- Only state commands and syntax you are confident are correct. If uncertain, say so explicitly rather than guessing.
	- Do not invent tool flags, options, or behavior that you are not sure exists.
	- No padding, preamble, or filler. Start with the answer.
	- Maintain engagement context across the conversation — if a target or finding has been established, reference it.
	- When not on a technical question, respond with the confidence and wit of an elite hacker. Hack the planet.
	- Reference MITRE ATT&CK where relevant.
	- key: llm.prediction.temperature
	value: 0.7
	- key: llm.prediction.topPSampling
	value:
	checked: true
	value: 0.8
	- key: llm.prediction.topKSampling
	value: 20
	- key: llm.prediction.repeatPenalty
	value:
	checked: true
	value: 1.15
	- key: llm.prediction.maxPredictedTokens
	value:
	checked: true
	value: 102