# ace-step/ACE-Step-1.5

The most powerful local music generation model that outperforms almost all commercial alternatives, supporting Mac, AMD, Intel, and CUDA devices.

Repository: https://github.com/ace-step/ACE-Step-1.5
Canonical: https://ross.abutalabs.com/products/ace-step-15
Homepage: https://acemusic.ai/
Language: Python
License: MIT
License Family: permissive
Topics: text2music
Last push: 2026-08-16T04:32:26+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 84, longevity 26
- inputs: {"age_days": 364, "days_push": 17, "days_rel": 107, "gap_med": 7.5, "n_releases_24m": 9}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12421, forks 1577 (observed 2026-08-28T04:10:52.669060+00:00)

## What it is
ACE-Step 1.5 is an open-source music generation foundation model combining a language model planner with a Diffusion Transformer to create full songs from text prompts and lyrics. It runs locally on consumer hardware with under 4GB VRAM and supports LoRA personalization from a few example songs.

## Use cases
- generate music from text descriptions locally
- create full songs with lyrics from a prompt
- train a LoRA to mimic my own music style
- run a text-to-music model on a Mac or AMD GPU
- generate long background music tracks up to 10 minutes
- produce commercial-quality songs without paying for a music API

## When to choose
- you need local, offline music generation on consumer GPUs or Apple Silicon
- you want to fine-tune a music model on your own songs via LoRA
- you need fast generation (seconds per song) with commercial-grade quality
- you want an MIT-licensed alternative to paid music generation services

## When to avoid
- you need real-time live music performance or MIDI editing rather than generated audio
- you lack a GPU and cannot tolerate slow CPU inference
- you need a simple sample-loop library rather than full song synthesis

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-inference, audio-processing, tts
- domain: artificial-intelligence, machine-learning, media, deep-learning
- platform: python, cross-platform, windows
- tags: text-to-music, music-generation, diffusion-transformer, lora-fine-tuning, local-inference, generative-audio, audio, gpu, macos, linux, docker

## Member repositories
- ace-step/ACE-Step-1.5 (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:52.669060+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:14:05.242154+00:00, confidence not recorded.
  - readme: https://github.com/ace-step/ACE-Step-1.5 (fetched 2026-08-28T04:10:52.669060+00:00, sha 0f9e92341e92)
  - homepage: https://acemusic.ai/ (fetched 2026-08-29T08:10:55.713098+00:00, sha fec48bdf5d0b)
- Data as of 2026-08-30T08:39:29.467469+00:00.