Instructions to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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": "Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M

Ollama
How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with Ollama:
```
ollama run hf.co/Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
```
Unsloth Desktop

How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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": "Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with Docker Model Runner:
```
docker model run hf.co/Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M
```

Lemonade

How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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 "Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-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"

Why this repository?

Unlike incomplete GGUF uploads, this repository provides the full 11-quantization spectrum (from high-precision F16 down to lightweight Q2_K) of the abliterated Qwen3-Coder-30B-A3B-Instruct model.

Choose the exact fit for your VRAM/RAM constraints without sacrificing reasoning capabilities.

Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated - GGUF

This is the integral and complete collection of quantizations in GGUF format for the Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated model, prepared locally for use with llama.cpp, Ollama, LM Studio, or Text-Generation-WebUI.

This model combines the advanced programming and logical reasoning capabilities of the Qwen3-Coder (MoE) architecture, but processed with abliteration techniques to root out artificial system censorship filters and blocks, responding without restrictions.

📋 Available Files (Complete Collection Without Splits)

File	Est. Size	BPW (Bits per Weight)	Recommended Usage Profile
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-F16.gguf`	~61.1 GB	16.00	Complete Base Template. Absolute fidelity of floating-point weights.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q8_0.gguf`	~32.5 GB	8.50	Identical quality to the original, ideal for maximizing performance in advanced local environments with high-end GPUs.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q6_K.gguf`	~25.1 GB	6.59	Extremely high retention of programming syntax and logic with an optimized weight.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q5_K_M.gguf`	~21.7 GB	5.69	Recommended Sweet Spot. Keeps code coherence intact while critically reducing weight.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q5_K_S.gguf`	~21.1 GB	5.54	Compact 5-bit variant.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q4_K_M.gguf`	~18.6 GB	4.85	The Most Wanted. Optimal balance for running software development inferences on advanced home setups.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q4_K_S.gguf`	~17.5 GB	4.58	Compact 4-bit variant to accelerate tokens-per-second speed.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q3_K_L.gguf`	~15.9 GB	4.01	Medium-high 3-bit compression. Retains basic programming logic.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q3_K_M.gguf`	~14.7 GB	3.66	Intermediate 3-bit variant.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q3_K_S.gguf`	~13.3 GB	3.44	Lightweight 3-bit variant.
`Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q2_K.gguf`	~11.3 GB	2.90	Extreme Compression. May experience structure degradation and complex code bleeding. For experimental development only.

Note: Sizes are initial baseline estimates based on the model's native weight in safetensors (31B MoE parameters); verifying the final size on disk after local compilation is recommended.

💡 Highlighted Usage

Since this model is optimized for coding tasks and uncensored instructions, you can call it directly in your terminal using llama.cpp:

./llama-cli -m Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated-Q4_K_M.gguf -n 2048 -p "Write a Python script to perform advanced web scraping while avoiding blocks."

If you prefer using Ollama, you can run it directly with:

ollama run huihui_ai/qwen3-coder-abliterated

⚖️ Disclaimer

This model is open-access and lacks artificial safety filters (abliterated). Its original alignment has been modified for experimental and research purposes. Generated content, whether text responses or software code, is the sole responsibility of the individual running the local inference.

Credits

Base Coder Model: Qwen / Alibaba Cloud (Qwen3-Coder-30B-A3B-Instruct)
Abliteration: huihui-ai
Complete GGUF Quantizations: Thaurock

Downloads last month: 406

GGUF

Model size

31B params

Architecture

qwen3moe

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for Thaurock/Qwen3-Coder-30B-A3B-Instruct-abliterated-GGUF

Base model

Qwen/Qwen3-Coder-30B-A3B-Instruct

Finetuned

huihui-ai/Huihui-Qwen3-Coder-30B-A3B-Instruct-abliterated

Quantized

(9)

this model