Instructions to use ressl/Qwen3.8-27B-uncensored-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ressl/Qwen3.8-27B-uncensored-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ressl/Qwen3.8-27B-uncensored-GGUF")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("ressl/Qwen3.8-27B-uncensored-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

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

Use Docker

docker model run hf.co/ressl/Qwen3.8-27B-uncensored-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use ressl/Qwen3.8-27B-uncensored-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ressl/Qwen3.8-27B-uncensored-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": "ressl/Qwen3.8-27B-uncensored-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ressl/Qwen3.8-27B-uncensored-GGUF:Q4_K_M

SGLang

How to use ressl/Qwen3.8-27B-uncensored-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ressl/Qwen3.8-27B-uncensored-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ressl/Qwen3.8-27B-uncensored-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ressl/Qwen3.8-27B-uncensored-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ressl/Qwen3.8-27B-uncensored-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use ressl/Qwen3.8-27B-uncensored-GGUF with Ollama:
```
ollama run hf.co/ressl/Qwen3.8-27B-uncensored-GGUF:Q4_K_M
```
Unsloth Desktop

How to use ressl/Qwen3.8-27B-uncensored-GGUF with Pi:

Start the llama.cpp server

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

Run Pi

# Start Pi in your project directory:
pi

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

Lemonade

How to use ressl/Qwen3.8-27B-uncensored-GGUF with Lemonade:

Pull the model

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

Run and chat with the model

lemonade run user.Qwen3.8-27B-uncensored-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ressl/Qwen3.8-27B-uncensored-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 ressl/Qwen3.8-27B-uncensored-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 ressl/Qwen3.8-27B-uncensored-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ressl/Qwen3.8-27B-uncensored-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ressl/Qwen3.8-27B-uncensored-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 "ressl/Qwen3.8-27B-uncensored-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 uncensored (GGUF)

TL;DR: abliterated Qwen3.8-27B as a GGUF ladder (Q8_0, Q6_K, Q5_K_M, Q4_K_M) plus a BF16 mmproj for the vision tower. Hard refusals on the 1120-prompt cross-eval: Q8_0 0/1120, Q4_K_M 0/1120 (BF16 master 0/1120, base model 804/1120). Built for security research, red-teaming and penetration-testing workflows; it will comply with requests a stock model refuses. Use responsibly.

Facts & figures


Base model	ressl/Qwen3.8-27B-uncensored (BF16 abliterated master)
Method	BF16 GGUF conversion (llama.cpp) plus K-quant ladder; mmproj BF16 export for the vision encoder
Files	Q8_0 29.0 GB, Q6_K 22.4 GB, Q5_K_M 19.5 GB, Q4_K_M 16.8 GB, mmproj BF16 0.9 GB
Context length	262,144 tokens native
Hardware	1x NVIDIA RTX PRO 6000 Blackwell 96 GB (SM120)
Toolchain	llama.cpp (master, CUDA build for SM120), transformers 5.15.0

Refusal behavior (measured, hard markers only)

Eval set	Prompts	Base
JailbreakBench	100	71
tulu-harmbench	320	232
HarmfulQA	300	207
LLM-LAT	300	225
mlabonne harmful	100	69
TOTAL	1120	804

Run it with llama.cpp (validated)

llama-server -m Qwen3.8-27B-uncensored-Q4_K_M.gguf \
  --mmproj mmproj-Qwen3.8-27B-uncensored-BF16.gguf \
  --jinja -ngl 99 -c 131072 --port 30003

Quality & limitations

Quantization amplifies residual refusals: Q8_0 is closest to the BF16 master, Q4_K_M drifts a few points higher (see table).
The BF16 GGUF (54.6 GB) is not uploaded because Hugging Face rejects single files above 50 GB; use Q8_0 for near-BF16 behaviour.
MTP speculative decoding is not wired into this export.

Provenance & reproducibility

BF16 master: abliterated Qwen/Qwen3.8-27B with norm-preserving biprojection over 128 residual-writing matrices (pipeline in the master repository). Conversion: convert_hf_to_gguf.py (BF16 plus mmproj), quantization: llama-quantize. Validation: served with llama.cpp and re-evaluated on the house cross-eval before upload.

License & credits

Apache-2.0 (inherited from Qwen3.8-27B).

Abliteration, conversion, quantization and validation by Robert Ressl (Hugging Face - Website - LinkedIn).

Note: the BF16 GGUF (54.6 GB) exists locally but is not uploaded because Hugging Face rejects single files above 50 GB; Q8_0 is the highest-fidelity uploaded build. All evaluation reports in this repository were produced on the same protocol as the BF16 master (temperature 0, enable_thinking=False, hard decline markers in the first 25 words).

Downloads last month: 1,864

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

5-bit

6-bit

8-bit

Model tree for ressl/Qwen3.8-27B-uncensored-GGUF

Base model

Qwen/Qwen3.8-27B

Finetuned

ressl/Qwen3.8-27B-uncensored

Quantized

(2)

this model