Instructions to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF",
	filename="Qwen3.6-35B-A3B-MXFP6.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
# Run inference directly in the terminal:
llama-cli -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
# Run inference directly in the terminal:
llama-cli -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
# Run inference directly in the terminal:
./llama-cli -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

Use Docker

docker model run hf.co/michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

LM Studio
Jan
Ollama
How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with Ollama:
```
ollama run hf.co/michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
```

Unsloth Studio

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF to start chatting

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

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": "michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

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 michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with Docker Model Runner:
```
docker model run hf.co/michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF
```

Lemonade

How to use michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull michaelw9999/Qwen3.6-35B-A3B-MXFP6-GGUF

Run and chat with the model

lemonade run user.Qwen3.6-35B-A3B-MXFP6-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

This is an proof of concept/work in progress Qwen3.6-35B-A3B quantized into MXFP6.
It was quantized with my still experimental advanced-gguf-quantizer tool.
This GGUF will ONLY work on llama.cpp. The CPU only PR is posted here:

https://github.com/ggml-org/llama.cpp/pull/22671

The PR runs very slowly because that is for the initial implementation without GPU support.

You may preview the very fast POC CUDA version from my fork:

https://github.com/michaelw9999/llama.cpp/tree/mxfp6-cuda

To merge into your existing llama.cpp installation:

git remote add mxfp6 https://github.com/michaelw9999/llama.cpp
git fetch mxfp6
git merge mxfp6/mxfp6-cuda
cmake -B build -DGGML_CUDA=ON
cmake --build build -j

Or to install fresh:

git clone -b mxfp6-cuda https://github.com/michaelw9999/llama.cpp
cd llama.cpp
cmake -B build -DGGML_CUDA=ON
cmake --build build -j

NOTICE:

This is my own work and is experimental and unofficial.

The CUDA version is not part of any llama.cpp PR (yet). This is not associated with NVIDIA in anyway.

Very likely, any future MXFP6 design will not be compatible with this implementation.

For Qwen3.6-35B, MXFP6 is almost as fast as NVFP4 on prefill.

Using FP8 for activations, it is faster than NVFP4 on tokengen.

Feedback is both requested and encouraged so I can make further improvements into future llama.cpp PRs.

The NVFP4/MXFP6 quantizer is still being improved and will be posted in the future. Please let me know if you want to see a specific model turned into MXFP6.

I will create an MTP enabled version soon.

MXFP6: Final estimate: PPL = 6.7890 +/- 0.04420

  Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, VRAM: 32606 MiB
| model                          |       size |     params | backend    | ngl |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | --------------: | -------------------: |
| qwen35moe 35B.A3B MXFP6 - E2M3 |  26.46 GiB |    34.66 B | CUDA       |  99 |           pp512 |      8094.43 ± 49.53 |
| qwen35moe 35B.A3B MXFP6 - E2M3 |  26.46 GiB |    34.66 B | CUDA       |  99 |           tg128 |        188.10 ± 3.20 |

  Device 0: NVIDIA GeForce RTX 5090, compute capability 12.0, VMM: yes, VRAM: 32606 MiB
| model                          |       size |     params | backend    | ngl |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | --------------: | -------------------: |
| qwen35moe 35B.A3B NVFP4        |  21.48 GiB |    34.66 B | CUDA       |  99 |           pp512 |      8220.18 ± 57.89 |
| qwen35moe 35B.A3B NVFP4        |  21.48 GiB |    34.66 B | CUDA       |  99 |           tg128 |        159.53 ± 0.82 |