Instructions to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with Transformers:

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

pipe = pipeline("image-text-to-text", model="tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF", device_map="auto")

llama-cpp-python

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF",
	filename="Qwen3.6-27B-16GB-VRAM-MTP-mini-IQ4_XS.gguf",
)

llm.create_chat_completion(
	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"
					}
				}
			]
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama cli -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama cli -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
# Run inference directly in the terminal:
./llama-cli -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

Use Docker

docker model run hf.co/tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

LM Studio
Jan

vLLM

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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": "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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/tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

SGLang

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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": "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 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 "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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": "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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"
						}
					}
				]
			}
		]
	}'

Ollama
How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with Ollama:
```
ollama run hf.co/tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
```

Unsloth Studio

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF to start chatting

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

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": "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

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 tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

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 "tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS" \
  --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"

Docker Model Runner
How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with Docker Model Runner:
```
docker model run hf.co/tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS
```

Lemonade

How to use tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF:IQ4_XS

Run and chat with the model

lemonade run user.Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF-IQ4_XS

List all available models

lemonade list

Qwen3.6-27B Mini - IQ4_XS (GGUF)

An optimally sized quantized version of Qwen3.6-27B was created to fit 16GB with usable MTP.

Model Details

Base Model: Qwen3.6-27B
Quantization: IQ4_XS (Mini)
BPW: 4.0012
Quantized by: Using Thireus' GGUF Tool Suite

Quick test

Wikitext-2-raw PPL: 7.0516 ± 0.04664
Decoding speed RTX4070 Ti Super : 80 t/s
Chessboard Test 😃

Hardware Requirements

Fully fits on 16GB VRAM with 92K context using MTP + q4_0 KV cache.
Can push even higher context by reducing KV cache further with TurboQuant or Kvarn.

Summary of tensor counts and bpw per qtype

QTYPE		Count	BPW	Assigned GiB	% Assigned	Max GiB (all)
+f32       	353	32    	  0.01 GiB	-		-
q8_0      	6  	8.5   	  0.00 GiB	 0.01%		26.61
q6_K      	101	6.5625	  0.06 GiB	 0.30%		20.55
q5_1      	0  	6     	  0.00 GiB	 0.00%		18.78
q5_K      	20 	5.5   	  0.08 GiB	 0.49%		17.22
q5_0      	0  	5.5   	  0.00 GiB	 0.00%		17.22
q4_1      	0  	5     	  0.00 GiB	 0.00%		15.65
q4_K      	0  	4.5   	  0.00 GiB	 0.00%		14.09
q4_0      	0  	4.5   	  0.00 GiB	 0.00%		14.09
iq4_nl    	0  	4.5   	  0.00 GiB	 0.00%		14.09
iq4_xs    	282	4.25  	  8.86 GiB	66.62%		13.31
q3_K      	0  	3.4375	  0.00 GiB	 0.00%		10.76
iq3_s     	89 	3.4375	  3.51 GiB	32.59%		10.76
iq3_xxs   	0  	3.0625	  0.00 GiB	 0.00%		9.59
q2_K      	0  	2.625 	  0.00 GiB	 0.00%		8.22
iq2_xs    	0  	2.3125	  0.00 GiB	 0.00%		7.24
iq2_xxs   	0  	2.0625	  0.00 GiB	 0.00%		6.46
iq1_m     	0  	1.75  	  0.00 GiB	 0.00%		5.48
iq1_s     	0  	1.5625	  0.00 GiB	 0.00%		4.92

Average BPW: 4.0012

Downloads last month: 5,462

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

Model tree for tooltd/Qwen3.6-27B-mini-IQ4-XS-MTP-16GB-VRAM-GGUF

Base model

Qwen/Qwen3.6-27B

Quantized

(647)

this model