🍎 macOS counterpart to NVIDIA DGX Spark Playbooks

Start Building on Mac

Find instructions and examples to run AI workloads on Apple Silicon.
Optimized for M-series · Unified Memory · Metal GPU Acceleration.

📚 29 Playbooks🗂️ 8 Categories⚡ Apple Silicon🔒 100% Local

Access Remotely

Connect to your Mac from anywhere

🌐

Set Up Remote Access

Tailscale creates a secure mesh VPN to access your Mac and its AI models from any device, anywhere.

Configure Now →

First Time Here?

Try these developer quickstarts

⚡

ComfyUI on macOS45 min

Node-based image generation workflow with MPS backend

→

⚡

FLUX / Stable Diffusion with MLX15 min

Generate images with FLUX and SD models natively on Mac

→

⚡

MLX LM for Inference10 min

Apple's high-performance native LLM engine — fastest on Mac

→

⚡

Ollama5 min

Install and run LLMs locally with a single command

→

⚡

Open WebUI with Ollama15 min

Deploy a full ChatGPT-like interface locally

→

What's New

Recently added playbooks

⚡

New⏱ 5 min

LM Studio on macOS

Polished GUI for local LLM inference using MLX backend

Cross-platform robotics simulation — replaces Isaac Sim/Lab

roboticssimulation

All Playbooks

29 playbooks

Step-by-step instructions for AI workloads on Apple Silicon

🚀onboarding

🚀

⏱ 5 min

Ollama

Install and run LLMs locally with a single command

inference

🚀

⏱ 15 min

Open WebUI with Ollama

Deploy a full ChatGPT-like interface locally

⚡inference

Run GGUF models with first-class Apple Silicon Metal acceleration

Polished GUI for local LLM inference using MLX backend

Apple's high-performance native LLM engine — fastest on Mac

Quantize any HF model to 2/4/8-bit for Apple Silicon

Run vision-language models locally with MLX

Serve models via OpenAI-compatible API with Ollama

Use draft models to accelerate generation 1.5–2.5×

inferenceoptimization

🧪fine tuning

🧪

⏱ 1 hr

LLaMA Factory on macOS

Install and fine-tune models with LLaMA Factory on MPS

fine-tuningui

🧪

⏱ 1 hr

FLUX LoRA Fine-tuning

Fine-tune FLUX.1 image models with LoRA on Mac

image generationfine-tuning

🧪

⏱ 30 min

MLX LoRA Fine-tuning

Fine-tune LLMs with LoRA/QLoRA natively on Apple Silicon

mlxfine-tuning

🧪

⏱ 1 hr

PyTorch MPS Fine-tuning

GPU-accelerated training using PyTorch Metal backend

📊data science

Run JAX with Metal GPU backend for scientific computing

jaxdata science

📊

⏱ 20 min

MLX + Accelerate Data Science

GPU-accelerated numerical computing on Apple Silicon

data sciencemlx

📊

⏱ 20 min

Portfolio Optimization

Convex optimization for portfolio allocation with cvxpy

data sciencefinance

📊

⏱ 15 min

Single-cell RNA Sequencing

End-to-end scRNA-seq workflow with scanpy

data sciencebioinformatics

🎨image generation

🎨

⏱ 45 min

ComfyUI on macOS

Node-based image generation workflow with MPS backend

image generationui

🎨

⏱ 15 min

FLUX / Stable Diffusion with MLX

Generate images with FLUX and SD models natively on Mac

🤖applications

Build a multi-agent system with LangGraph + local Ollama

agentslangchain

🤖

⏱ 30 min

RAG with LangChain + Ollama

Fully local retrieval-augmented generation pipeline

raglangchain

🤖

⏱ 30 min

Text to Knowledge Graph

Extract knowledge triples with local LLMs and graph DBs

knowledge graphnlp

🤖

⏱ 30 min

Video Search & Summarization

Transcribe and analyze video with Whisper + VLMs

🦾robotics

Cross-platform robotics simulation — replaces Isaac Sim/Lab

roboticssimulation

🔧tools

🔧

⏱ 5 min

System Monitoring (asitop)

Monitor GPU, CPU, ANE, and memory on Apple Silicon

toolsmonitoring

🔧

⏱ 5 min

macOS Network Sharing

SSH, screen sharing, and remote access setup

Secure mesh networking for remote access

toolsnetworking

🔧

⏱ 15 min

Vibe Coding with Continue.dev

Local AI coding assistant with Ollama + VS Code

Install and configure VS Code for ML development on Mac

tools

🍎mac-playbooks

macOS counterpart to NVIDIA DGX Spark Playbooks. Step-by-step AI/ML playbooks for Apple Silicon.

Resources

GitHub MLX Docs Ollama DGX Spark Playbooks