Instructions to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit 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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
# Run inference directly in the terminal:
llama cli -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
# Run inference directly in the terminal:
llama cli -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

Use Docker

docker model run hf.co/TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

LM Studio
Jan
Ollama
How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with Ollama:
```
ollama run hf.co/TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
```

Unsloth Studio

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit 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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit 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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit to start chatting

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

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": "TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

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 "TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0" \
  --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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with Docker Model Runner:
```
docker model run hf.co/TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0
```

Lemonade

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

Run and chat with the model

lemonade run user.mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit-Q8_0

List all available models

lemonade list

Hermes Agent

How to use TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

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 TracNetwork/mayhem-catalog-SulphurAI-Sulphur-2-base-MLX-4bit:Q8_0

Run Hermes

hermes

Atomic Chat

OpenMayhem Sulphur-2 Distilled MLX Bundle

This is the admin-published canonical MLX artifact bundle for SulphurAI/Sulphur-2-base.

Sulphur artifact: MLXBits/sulphur-2-distill-mlx-q4 at d210a0937cac3464ef80c74806e886beddf19a8e
Text encoder: mlx-community/gemma-3-12b-it-4bit at 86cc6a8dedbc456dd0e4af01a9d09f396f77e558
Sulphur source: SulphurAI/Sulphur-2-base at 875e886e556b955d21149316fd631cc121db6cc1
Runtime: dgrauet/ltx-2-mlx at e1838a855bfd1640135c424c96cb27a0c0ad150e

The complete byte inventory is bound by mayhem-sulphur-mlx-runtime.json. The optional uncensored prompt enhancer is pinned at the same Sulphur source revision as a Q8 GGUF plus its BF16 projector. OpenMayhem runs that pair through its verified llama.cpp compositor only when a provider advertises the capability and a request explicitly opts in. Serving requires OpenMayhem's signed catalog entry and real joint audio/video canary evidence.

The Sulphur/LTX weights and derivatives are governed by the included LICENSE-LTX-2.3 terms. The Gemma text encoder remains subject to the Gemma terms referenced by its included model card.

Downloads last month: 14

GGUF

Model size

9B params

Architecture

qwen35

Hardware compatibility

8-bit