Instructions to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

Use Docker

docker model run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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": "SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

Ollama
How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with Ollama:
```
ollama run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
```

Unsloth Studio

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF to start chatting

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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": "SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with Docker Model Runner:
```
docker model run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M
```

Lemonade

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.8-27B-SwissNeuron-Derisked-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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 "SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-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"

Qwen3.8-27B-SwissNeuron-Derisked-GGUF

llama.cpp GGUF releases of SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked, built from the repaired SFT plus latest-SFT DWM α=0.1 BF16 checkpoint.

Files

Quant	Use
BF16	Lossless GGUF reference
Q8_0	Maximum quantized quality
Q6_K	Near-BF16 quality
Q5_K_M	High-quality balanced option
Q4_K_M	Recommended general-purpose size/quality option

Exact sizes and SHA-256 hashes are in SHA256SUMS.

The repository also includes:

mmproj-Qwen3.8-27B-SwissNeuron-Derisked-BF16.gguf — BF16 vision projector
mtp-Qwen3.8-27B-SwissNeuron-Derisked-BF16.gguf — standalone BF16 MTP draft head

llama.cpp

Use a recent llama.cpp build with Qwen3.8 / qwen35 Gated-DeltaNet support.

llama-server \
  -hf SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M \
  -c 16384 --port 8080

Thinking is supported by the embedded chat template. For non-thinking mode, pass the current llama.cpp reasoning-off option or the equivalent chat-template kwargs supported by your frontend.

LM Studio

Search for SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF inside LM Studio. Select the BF16 file for the native, lossless derisked SwissNeuron checkpoint, or Q4_K_M/Q5_K_M/Q6_K/Q8_0 for smaller deployments. Keep the embedded Qwen3.8 chat template.

Ollama

Run directly from Hugging Face with a recent Ollama release:

# Native lossless BF16 GGUF
ollama run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:BF16

# Recommended smaller quant
ollama run hf.co/SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF:Q4_K_M

Or download the repository's Modelfile and run:

ollama create swissneuron-qwen38-bf16 -f Modelfile
ollama run swissneuron-qwen38-bf16

Notes

These files contain the language model and native MTP tensors supported by current llama.cpp.
Vision requires a compatible Qwen3.8 mmproj file; this initial release is verified for text inference.
The source HF model advertises factor-4 YaRN to 1M, but practical GGUF context is constrained by host/GPU memory and runtime state. Start at 16K–64K and scale only after retrieval testing.

Downloads last month: 281

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked-GGUF

Base model

Qwen/Qwen3.8-27B

Finetuned

SwissNeuron/Qwen3.8-27B-SwissNeuron-Derisked

Quantized

(4)

this model