Run ACE-Step 1.5 Locally on Mac — Free AI Music Generation in Minutes

Run ACE-Step 1.5 Locally on Mac — Free AI Music Generation in Minutes

Deploy the ACE-Step 1.5 AI music generation model entirely offline on your Mac. Apple Silicon optimized, no subscription required — just a few terminal commands.

Since so many of you are curious about running AI locally, here’s a fun one — ACE-Step 1.5.

In short, it lets you generate music right on your own Mac, completely free, no internet required, and the project has solid Apple Silicon (MPS) optimizations baked in.

Don’t worry if you’re not a developer — the whole setup takes just a handful of commands. Follow the steps below and you’ll have it running in no time.

Setup

Open your Terminal and let’s get started.

1. Choose a home for the project

Pick a directory you’re comfortable with and navigate to it.

cd Projects
cd Python

2. Clone the repo & install dependencies

Clone the project from GitHub, then use uv to install dependencies quickly. If you haven’t tried uv yet, it’s highly recommended — it’s a fast, modern Python environment manager.

git clone https://github.com/ACE-Step/ACE-Step-1.5.git
cd ACE-Step-1.5
uv sync

Wait for the progress bar to finish before moving on.

3. Verify Apple Silicon acceleration

Before running the model, confirm that your Apple chip (MPS backend) is recognized correctly.

uv run python -c "import torch; print(f'MPS check: {torch.backends.mps.is_available()}')"

If the terminal prints MPS check: True, everything is good — your GPU is ready to work.

Running the App

4. Start the service

Simply run:

uv run acestep

5. Open the web UI

Once the terminal shows it’s running, open your browser and go to: 127.0.0.1:7860

You’ll see the ACE-Step interface.

A few important things to know:

  • Choose the right memory config: The UI asks you to select your model VRAM size. Match it to your Mac’s Unified Memory — for example, select 16-20GB if that’s what you have.
  • Initialize the service: Click “Initialize Service”.
    • ⚠️ Note: The first run will automatically download the model weights, which are fairly large. The UI may appear frozen for a while — this is normal. Grab a coffee and wait for it to finish.

6. Generate your first AI track

Once initialized, it’s surprisingly simple:

  1. Switch to “Simple” mode.
  2. Type a prompt — if you’re not sure what to write, try easy example.
  3. Click “Create Sample”. The “Custom” parameters below will be filled in automatically — ignore them for now.
  4. Hit “Generate Music” at the bottom.

That’s it! Wait for the progress bar to complete, and you’ll hear music generated entirely on your local machine 🎵