Instructions to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Transformers:

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

pipe = pipeline("text-generation", model="electroglyph/Qwen3-4B-Instruct-2507-uncensored")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("electroglyph/Qwen3-4B-Instruct-2507-uncensored")
model = AutoModelForCausalLM.from_pretrained("electroglyph/Qwen3-4B-Instruct-2507-uncensored")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="electroglyph/Qwen3-4B-Instruct-2507-uncensored",
	filename="GGUF/Qwen3-4B-Instruct-2507-uncensored-UD-Q4_K_XL.gguf",
)

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

Inference
Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
# Run inference directly in the terminal:
llama-cli -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
# Run inference directly in the terminal:
llama-cli -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

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 electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
# Run inference directly in the terminal:
./llama-cli -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

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 electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

Use Docker

docker model run hf.co/electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

LM Studio
Jan

vLLM

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "electroglyph/Qwen3-4B-Instruct-2507-uncensored"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "electroglyph/Qwen3-4B-Instruct-2507-uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

SGLang

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored 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 "electroglyph/Qwen3-4B-Instruct-2507-uncensored" \
    --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": "electroglyph/Qwen3-4B-Instruct-2507-uncensored",
		"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 "electroglyph/Qwen3-4B-Instruct-2507-uncensored" \
        --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": "electroglyph/Qwen3-4B-Instruct-2507-uncensored",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Ollama:
```
ollama run hf.co/electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
```

Unsloth Studio

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored 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 electroglyph/Qwen3-4B-Instruct-2507-uncensored 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 electroglyph/Qwen3-4B-Instruct-2507-uncensored to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for electroglyph/Qwen3-4B-Instruct-2507-uncensored to start chatting

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

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": "electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

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 electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

Run Hermes

hermes

Docker Model Runner
How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Docker Model Runner:
```
docker model run hf.co/electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL
```

Lemonade

How to use electroglyph/Qwen3-4B-Instruct-2507-uncensored with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull electroglyph/Qwen3-4B-Instruct-2507-uncensored:UD-Q4_K_XL

Run and chat with the model

lemonade run user.Qwen3-4B-Instruct-2507-uncensored-UD-Q4_K_XL

List all available models

lemonade list

This is a simple SFT finetune of Qwen3-4B-Instruct-2507 to remove censorship.

I've tried to do the bare minimum training necessary to leave the model as pristine as possible.

I'll follow this up with an "unslop" version that attempts to mitigate some of the slop.

edit: next model is here

I've uploaded a UD-Q4_K_XL GGUF with settings that I grabbed from Unsloth's quant using my lil utility: quant_clone

Downloads last month: 58

Safetensors

Model size

4B params

Tensor type

BF16