Instructions to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF",
	filename="Nvidia-Qwen3.6-27B-NVFP4-A.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
# Run inference directly in the terminal:
llama cli -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
# Run inference directly in the terminal:
llama cli -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
# Run inference directly in the terminal:
./llama-cli -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
# Run inference directly in the terminal:
./build/bin/llama-cli -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

Use Docker

docker model run hf.co/CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

LM Studio
Jan

vLLM

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CodeFault/Nvidia-Qwen3.6-27B-NVFP4-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": "CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

Ollama
How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with Ollama:
```
ollama run hf.co/CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
```

Unsloth Studio

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF to start chatting

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

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": "CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

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 "CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4" \
  --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 CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with Docker Model Runner:
```
docker model run hf.co/CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4
```

Lemonade

How to use CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4

Run and chat with the model

lemonade run user.Nvidia-Qwen3.6-27B-NVFP4-GGUF-NVFP4

List all available models

lemonade list

Nvidia-Qwen3.6-27B-NVFP4 - GGUF

Quantized GGUF versions of nvidia/Qwen3.6-27B-NVFP4. These were generated using llama.cpp's convert_hf_to_gguf.py (b9859).

Nvidia-Qwen3.6-27B-NVFP4-A.gguf - All layers are NVFP4 quantized. This required modifying convert_hf_to_gguf.py, and needs cleaning up before possible upstreaming.
Nvidia-Qwen3.6-27B-NVFP4-BF16-Attn.gguf: NVFP4 FFN layers are preserved, while FP8 attention layers are upcasted to BF16. This is the default conversion for BF16 because GGUF files do not support FP8.

Quantizations provided

File	Quantization	Size
Nvidia-Qwen3.6-27B-NVFP4-A.gguf	NVFP4	17.9 GB
Nvidia-Qwen3.6-27B-NVFP4-BF16-Attn.gguf	NVFP4 FFN, BF16 attention	28.2 GB

Perplexity test

I tested perplexity using llama-perplexity and Salesforce's wikitext-2-raw-v1.

File	Ctx	PPL
Nvidia-Qwen3.6-27B-NVFP4-A.gguf	512	7.7540 ± 0.05396
Nvidia-Qwen3.6-27B-NVFP4-BF16-Attn.gguf	512	7.4814 ± 0.05157

Evaluation

The following models were evaluated for a fair comparison of capability, size and speed.

Model	Quantization	Size	Reason
unsloth/Qwen3.6-27B-MTP-GGUF	UD-Q4_K_XL	17.9 GB	Closest non-NVFP4 in size to NVFP4.
unsloth/Qwen3.6-27B-MTP-GGUF	UD-Q6_K_XL	26 GB	Closest non-NVFP4 in size to BF16-Attn.
unsloth/Qwen3.6-27B-NVFP4	NVFP4¹	25.4 GB	Alternative NVFP4 quant.

1: unsloth/Qwen3.6-27B-NVFP4 does not provide a GGUF. I used llama.cpp's conversion which passes through Unsloth's NVFP4 tensors.

	CodeFault NVFP4	CodeFault BF16-Attn	Unsloth NVFP4	Unsloth UD-Q4_K_XL	Unsloth UD-Q6_K_XL
Coding
HumanEval	0.8415 ± 0.0286	0.8354 ± 0.029	0.811 ± 0.0307	0.8354 ± 0.029	0.8537 ± 0.0277
HumanEval+	0.7866 ± 0.0321	0.7927 ± 0.0318	0.7744 ± 0.0327	0.7805 ± 0.0324	0.7805 ± 0.0324
MBPP	0.006 ± 0.0035^!!	0.754 ± 0.0193	0.742 ± 0.0196	0.756 ± 0.0192	0.754 ± 0.0193
MBPP+	0.0106 ± 0.0053^!!	0.8836 ± 0.0165	0.8995 ± 0.0155	0.8968 ± 0.0157	0.8836 ± 0.0165
Instruction
IFEval			0.8447 ± 0.0156	0.841 ± 0.0157	0.8447 ± 0.0156
Knowledge
ARC-Challenge	0.9659 ± 0.0053	0.971 ± 0.0049	0.971 ± 0.0049	0.971 ± 0.0049	0.971 ± 0.0049
MMLU-Pro				0.835 ± 0.0033
STEM & Reasoning
BIG-Bench Hard				0.926 ± 0.003
GPQA Diamond
GSM8K			0.9098 ± 0.0079	0.9083 ± 0.008	0.9158 ± 0.0076
Hendrycks Math

NOTICE: These tests are actively running.

!!: Such a drastic failure suggests something is wrong with the harness, not the model. I still need to investigate.

These evaluations were run using lm_eval. The models were run in instruct (non-thinking) mode with the following parameters in llama-server (b9775):

ctx-size = 32768
cache-type-k = q8_0
cache-type-v = q8_0
top-k = 20
top-p = 0.8
min-p = 0
presence-penalty = 1.5
spec_type = draft-mtp
spec_draft_n_max = 2
chat-template-kwargs = {"enable_thinking":false}

Benchmarks

Benchmarks: Coming after evaluations.

Serving with llama.cpp

It has a max context size of 262,114. This can be served using:

llama-server \
-hf CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF:NVFP4 \
--temp 0.6 \
--top-p 0.95 \
--top-k 20 \
--repeat-penalty 1.1 \
--spec-type draft-mtp \
--spec-draft-n-max 2

Downloads last month: 920

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

16-bit

Model tree for CodeFault/Nvidia-Qwen3.6-27B-NVFP4-GGUF

Base model

Qwen/Qwen3.6-27B

Quantized

nvidia/Qwen3.6-27B-NVFP4

Quantized

(3)

this model