ACE-Step

ACE-Step is an open-source AI music generation tool that creates full songs with vocals, instrumentals, and lyrics using a Diffusion Transformer, supporting…

  • AI
  • Media Applications
  • Tools & Utilities
ACE-Step is an open-source AI music generation tool that creates full songs with vocals, instrumentals, and lyrics using a Diffusion Transformer, supporting text-to-music, remixing, cover generation, and LoRA fine-tuning.

Platform availability

Available on 1 of 9 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

69/100 Good

Maintained
75last commit 74 days ago
Popular
8314,640 GitHub stars
Easy to install
351 install platform, install notes, Docker image
Light to run
No data
Security check
76no serious findings, 4 minor, CI, tests

Based on 4 of 5 factors.

Availability: platform listings confirmed today, links checked 28 days ago

GitHub stars
14,640
Open issues
163
Last commit
2026-08-29
Latest release
v0.1.8
License
MIT
Activity
stale

Checked today - source: GitHub (ace-step/ACE-Step-1.5)

Resources & Compatibility

Minimum requirements: 10 GB disk. ARM (e.g. Raspberry Pi) is not supported.

Checked 43 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Use a supported system Before installing, make sure you are on Linux, Windows, or Android. Those are the systems the get-started guide lists, and using an unsupported platform can stop the app from working properly. source
  • Pick your stream format When starting a broadcast, choose whether you want to start an MPEG TS or HLS broadcast. MPEG TS and HLS are different video-streaming formats, and the wrong choice can mean players or viewers cannot receive the stream correctly. source
  • Set up nodes and trackers If you are broadcasting, review the source and support node setup plus tracker options. These parts help Ace Stream find the stream and share it, so missing or misconfigured items can make your broadcast unavailable or slow. source
  • First Run Models (~10 GB) are downloaded automatically on first start, which may take several minutes depending on your internet connection. Subsequent starts are instant. source
  • GPU Requirements & Model Selection Requires an NVIDIA GPU with CUDA support. DiT models: turbo (default, 8 steps, fastest), turbo-rl (RL-refined), sft (50 steps, higher quality), base (50 steps, all features). Language model: 1.7B (default, 12–16 GB VRAM recommended) or 0.6B (GPUs under 12 GB VRAM). source

Fetched 28 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

How much model data will ACE-Step download on first launch?

ACE-Step automatically downloads about 10GB of models on first start, which can take several minutes depending on internet speed. Subsequent starts are instant.

Which ACE-Step DiT model should I pick for faster or higher-quality output?

Use turbo for fast 8-step generation, turbo-rl for turbo with reinforcement-learning refinement, sft for higher-quality 50-step generation, or base for 50-step generation with all features.

Will every ACE-Step feature work through ComfyUI?

A user reports that some advertised ACE-Step features seem missing when using ACE-Step 1.5 through ComfyUI. Verify the specific features you need before relying on the ComfyUI workflow.

What common ACE-Step/ComfyUI_ACE-Step problems does troubleshooting documentation cover?

The troubleshooting page covers installation problems, model loading errors, hardware constraints, generation failures, and performance issues.

What does the DGX Spark ACE-Step 1.5 forum post cover?

An NVIDIA forum post shares an ACE-Step v1.5 install recipe for DGX Spark. It includes measured performance numbers and scheduling observations for a production-style setup.

Answers sourced from deepwiki.com, forum.level1techs.com, forums.developer.nvidia.com, github.com

Community

Collected 42 days ago - public community threads