ACE-Step 1.5
Meet ACE-Step 1.5, an on-device music model that beats Suno on common eval metrics
FreeAI10,965 impressions#1 of its week
ACE-Step 1.5 is an open-source on-device music model that generates full songs in seconds on a local PC.
- for
- Developers and creators who need fast, local AI-generated music.
- pricing
- open source
- license
- MIT
Key features
- Full-song generation — Creates an entire song in about 2 seconds on an A100 GPU.
- Low VRAM requirement — Runs on a typical PC with roughly 4 GB of VRAM.
- Open-source model — Source code and weights are publicly available for modification.
- Higher evaluation scores — Outperforms Suno on common music evaluation metrics.
- ACE Studio integration — Provides an all-in-one AI music studio for editing and exporting tracks.
Use cases
- Rapid prototyping of background music for games or videos
- Generating custom jingles for marketing campaigns
- Embedding AI-generated music in desktop applications
- Creating full-length tracks for independent artists without cloud services
ACE-Step 1.5 FAQ
What is ACE-Step 1.5?+
ACE-Step 1.5 is an open-source music generation model that can produce a complete song in seconds on consumer-grade hardware.
Can it run locally?+
Yes, it runs on a typical PC with about 4 GB of VRAM, eliminating the need for cloud inference.
What hardware is required for optimal performance?+
On an NVIDIA A100 GPU it generates a song in ~2 seconds; a standard PC with 4 GB VRAM can also run the model, though slower.
How does its quality compare to other models?+
It beats Suno on common evaluation scores, indicating higher perceived quality.
Where can I access the code and model?+
The project is open source and available via the ACE Music GitHub repository linked from the website.
Summarized by DevHunt from acemusic.ai · Sep 27, 2026. Details may change; check the official site.
