⚡Ultra-Fast GenerationUnder 2s per full song on A100, under 10s on RTX 3090. Supports 10s to 10min audio.
🎵Commercial-Grade Quality1000+ instruments and styles, 50+ languages, fine-grained timbre description and metadata control.
🎛Versatile EditingCover generation, repainting, track separation, vocal-to-BGM, multi-track generation, and more.
🚀Lightweight & CustomizableRuns locally with <4GB VRAM. Train a LoRA from just a few songs to capture your own style.