Instructions to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
# Run inference directly in the terminal:
llama cli -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
# Run inference directly in the terminal:
llama cli -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
# Run inference directly in the terminal:
./llama-cli -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
# Run inference directly in the terminal:
./build/bin/llama-cli -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

Use Docker

docker model run hf.co/GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

LM Studio
Jan

vLLM

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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": "GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

Ollama
How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with Ollama:
```
ollama run hf.co/GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
```

Unsloth Studio

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF to start chatting

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

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": "GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

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 "GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL" \
  --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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with Docker Model Runner:
```
docker model run hf.co/GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL
```

Lemonade

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

Run and chat with the model

lemonade run user.Qwen3.6-27B-IQ4_NL-with-MTP-GGUF-IQ4_NL

List all available models

lemonade list

Hermes Agent

How to use GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

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 GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL

Run Hermes

hermes

Atomic Chat

Qwen3.6-27B IQ4_NL with MTP GGUF

This GGUF combines a smart-mix IQ4_NL quantization of Qwen3.6-27B with the MTP (Multi-Token Prediction) head grafted from the unsloth IQ4_XS MTP build, enabling native speculative decoding at a compact file size. Optimized for 16 GB VRAM GPUs (RTX 4060 Ti 16GB, RTX 4070 Ti Super, RX 7800 XT, etc.).

File

File	Size	Quantization
`qwen3.6-27b-IQ4_NL-with-MTP.gguf`	~14.19 GB	Smart mix: IQ4_NL + Q4_K + IQ4_XS + F32

Quantization Strategy

This is not a pure quantization. It uses a smart mix that assigns different precision levels based on tensor sensitivity:

Component	Type	Precision	Rationale
Critical layers (FFN gate/down, SSM)	IQ4_NL	High	Non-linear reconstruction preserves quality on sensitive weights
Attention QKV projections	Q4_K	Medium	K-quant with per-group scales, good balance for attention
FFN up projections	IQ4_XS	Aggressive	Less sensitive, can tolerate tighter compression
MTP attention/FFN	Q4_K	Medium	Draft head needs reasonable precision for acceptance
MTP FFN up	IQ4_XS	Aggressive	Same rationale as body
MTP prediction head (eh_proj)	Q8_0	Highest	Token prediction is the most sensitive — kept at full 8-bit
All norm tensors	F32	Full	1D tensors are tiny, no benefit to quantizing

What's Inside

Body (851 tensors): Smart-mix quantization using the unsloth imatrix for calibration. Critical FFN and SSM tensors at IQ4_NL, attention projections at Q4_K, FFN up projections at IQ4_XS.
MTP head (15 tensors): Extracted from unsloth/Qwen3.6-27B-MTP-GGUF (IQ4_XS variant), preserving the original mixed quantization for optimal speculative decoding acceptance.

Tensor Type Distribution

Type	Body Count	MTP Count	Total
IQ4_NL	370	0	370
Q4_K	64	6	70
IQ4_XS	64	1	65
Q8_0	0	1	1
F32	353	7	360
Total	851	15	866

Why This Exists

The Ununnilium pure GGUF strips MTP tensors to save space, but that means speculative decoding can't use the trained native draft head. This file grafts the MTP head back in, restoring native MTP speculative decoding while keeping a smart quantization mix that prioritizes quality where it matters most.

Provenance

Base model: Qwen/Qwen3.6-27B
Body quantization: Ununnilium/Qwen3.6-27B-IQ4_XS-pure-GGUF
MTP head source: unsloth/Qwen3.6-27B-MTP-GGUF (IQ4_XS variant)

Quickstart

llama.cpp

llama-server -hf GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF:IQ4_NL \
  --spec-type draft-mtp \
  --spec-draft-n-max 3 \
  --spec-draft-p-min 0.75

LM Studio

Search for GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF and load the file. Enable speculative decoding in settings with draft-mtp type.

llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF",
    filename="qwen3.6-27b-IQ4_NL-with-MTP.gguf",
    n_gpu_layers=-1,
    n_ctx=131072,
    speculative="draft-mtp",
    speculative_n_draft=3,
)

Recommended Server Flags

llama-server \
  --model qwen3.6-27b-IQ4_NL-with-MTP.gguf \
  --ctx-size 131072 \
  --n-gpu-layers 99 \
  --parallel 1 \
  --batch-size 2048 \
  --ubatch-size 128 \
  --cache-type-k q4_0 \
  --cache-type-v q4_0 \
  --threads 6 \
  --threads-batch 12 \
  --flash-attn on \
  --no-mmap \
  --spec-type draft-mtp \
  --spec-draft-n-max 3 \
  --spec-draft-p-min 0.75 \
  --temp 0.8 \
  --top-p 0.95 \
  --top-k 20 \
  --min-p 0.0

MTP Performance

Tested on AMD Radeon RX 7800 XT (16 GB VRAM) with Vulkan backend:

Metric	Value
Cumulative token acceptance	~82%
Draft acceptance rate	~92%
Generation speed	~42-65 t/s (varies by workload)
Prompt processing	~100-120 t/s

Acceptance rates above 60% mean MTP provides a net speedup. The ~82% cumulative token acceptance observed in testing indicates strong speculative decoding performance.

Model Architecture

Parameter	Value
Architecture	qwen35
Parameters	27B
Layers	64 + 1 MTP
Hidden size	5120
FFN size	17408
Attention heads	24 Q / 4 KV
Context length	262,144 (native)
Vocab size	248,320

Caveats

MTP support requires llama.cpp build with draft-mtp speculative decoding support (PR #22673 or newer)
The MTP head was trained against the original Qwen3.6-27B trunk; acceptance may vary if the body has been further fine-tuned
Vision/MTP combination has been fragile in llama.cpp testing; use text-only first
--parallel 1 is required when using MTP (parallel slots not yet supported)

Downloads last month: 221

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

Model tree for GianniDPC/Qwen3.6-27B-IQ4_NL-with-MTP-GGUF

Base model

Qwen/Qwen3.6-27B

Quantized

(682)

this model