ACE-Step
ACE Studio + StepFun music foundation model — 4 minutes of music from 20 seconds
The ACE Studio + StepFun music foundation model: generates 4 minutes of music from 20 seconds.
Generate music and songs with open AI models.
Key features
- Open-source music generation foundation model, 3.5B params
- Synthesizes up to 4 minutes of music in 20 seconds on A100
- 19 languages; diverse genres and vocal styles
- Controls: voice cloning, lyric editing, remixing, track generation
- Runs in 8GB VRAM with memory optimizations; ComfyUI node
Good for
- AI music creation for artists and producers
- Text-to-music with lyric alignment
Stack & tags
Pythonmusic-generationtext-to-musicopen-weightsdiffusionlyricsvoice-cloningcomfyui
Quick start
acestep --torch_compile true --cpu_offload true --overlapped_decode trueFrom the project README on GitHub.
Similar open-source projects
ACE-Step-1.5
★ 13,569The most powerful local music generation model that outperforms almost all commercial alternatives, supporting Mac, AMD, Intel, and CUDA devices.
Magenta RealTime
★ 1,807Google Magenta's first real-time music generation open model