Instructions to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF", dtype="auto")

llama-cpp-python

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF",
	filename="Qwen3.6-27B.i1-IQ4_XS-attn_qkv-IQ4_XS.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
# Run inference directly in the terminal:
llama-cli -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
# Run inference directly in the terminal:
./llama-cli -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

Use Docker

docker model run hf.co/cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

LM Studio
Jan

vLLM

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cHunter789/Qwen3.6-27B-i1-IQ4_XS-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": "cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

SGLang

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 "cHunter789/Qwen3.6-27B-i1-IQ4_XS-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": "cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 "cHunter789/Qwen3.6-27B-i1-IQ4_XS-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": "cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Ollama:
```
ollama run hf.co/cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
```

Unsloth Studio

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF to start chatting

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-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": "cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf cHunter789/Qwen3.6-27B-i1-IQ4_XS-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 cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

Run Hermes

hermes

Docker Model Runner
How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Docker Model Runner:
```
docker model run hf.co/cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS
```

Lemonade

How to use cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF:IQ4_XS

Run and chat with the model

lemonade run user.Qwen3.6-27B-i1-IQ4_XS-GGUF-IQ4_XS

List all available models

lemonade list

Qwen3.6-27B-i1-IQ4_XS (Fully Optimized)

EDIT (01.05.2026)

The mentioned regression is fixed (https://github.com/ggml-org/llama.cpp/issues/22544). This give us a hope for better & smaller IQ_4XS in the future!

Motivation

Recent updates in the llama.cpp repository (specifically commit 1dab5f5a44) introduced a hardcoded minimum quantization of q5_K for attn_qkv layers. While this was likely intended to preserve model quality, it causes a noticeable bloat in the final file sizes.

For comparison, the highly efficient Qwen3.5-27B iq4_xs by mradermacher weighed in at 14.7GB, whereas the equivalent Qwen3.6 i1-GGUF under the new commit rules swelled to over 15.1GB.

Methodology

To restore the optimal balance of size and performance, I modified the llama.cpp source code to revert the quantization of attn_qkv layers back to a pure IQ4_XS format. This mirrors the exact 1:1 layer quantization strategy originally used in mradermacher's Qwen3.5-27B release.

This model was quantized utilizing the imatrix provided by mradermacher: Qwen3.6-27B-i1-GGUF.

Performance vs. Size Trade-off

Extensive perplexity testing (llama-perplexity with pg19.txt, 65k context, Q8_0 cache) confirms that forcing IQ4_XS across all attn_qkv layers results in a statistically insignificant intelligence drop (+0.0039 PPL) while noticeably reducing the memory footprint.

./llama-perplexity -m Qwen3.6-27B.i1-IQ4_XS.gguf -f pg19.txt -c 65536 --chunks 32 -ngl -1 -ctk q8_0 -ctv q8_0 -fa 1 -b 512 -ub 128

./llama-perplexity -m Qwen3.6-27B.i1-IQ4_XS-attn_qkv-IQ4_XS.gguf -f pg19.txt -c 65536 --chunks 32 -ngl -1 -ctk q8_0 -ctv q8_0 -fa 1 -b 512 -ub 128

🧠 Intelligence (Perplexity) Comparison

Model Version	Perplexity (PPL)	Difference / Quality Drop
Standard `IQ4_XS` (with q5_K attn_qkv)	7.3765 ± 0.02760	Baseline
Custom `IQ4_XS` (attn_qkv iq4)	7.3804 ± 0.02762	+ 0.0039 (Negligible)

Conclusion: By utilizing this custom build, users save 375 MiB of active memory and reduce the static file size closer to the 14.7GB mark, with a practically non-existent impact on output quality (~0.05% PPL variance).

Downloads last month: 7,928

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

Model tree for cHunter789/Qwen3.6-27B-i1-IQ4_XS-GGUF

Base model

Qwen/Qwen3.6-27B

Quantized

(416)

this model