Run This Ai
EN DE

Demucs

Meta's AI music source separation — split any song into vocals, drums, bass, and other stems

★ 10,263 GitHub MIT music-separationstem-separationvocal-isolationsource-separationaudiometa Image & Video

Overview

Demucs is Meta's state-of-the-art open-source music source separation tool. With over 10,000 GitHub stars, it uses a hybrid spectrogram and waveform model to split any mixed audio track into individual stems: vocals, drums, bass, and other instruments. Demucs v4 (Hybrid Transformer) achieves the best separation quality available in open-source, rivaling commercial solutions. It supports batch processing of multiple files, real-time separation on GPU, and can handle both music and general audio. The tool includes a simple Python API and CLI, processes audio in seconds on GPU, and works on CPU for slower but functional operation. Demucs is widely used by musicians, DJs, and content creators for creating karaoke tracks, isolating vocals, and remixing.

Requirements

Min vCPU
2
Min RAM
4096 MB
Min Disk
10 GB
Rec vCPU
4
Rec RAM
8192 MB
Rec Disk
20 GB

Recommended VPS

Hostinger · KVM 4

4 vCPU · 16384 MB · 200 GB

$11.99
View plan

Hostinger · KVM 4

4 vCPU · 16384 MB · 200 GB

$11.99
View plan

Hostinger · KVM 8

8 vCPU · 32256 MB · 400 GB

$20.00
View plan

Affiliate disclosure

Docker Compose

# Generated by Run This Ai — docker-compose.yml
services:
  demucs:
    image: joanfont/demucs:latest
    restart: unless-stopped
    ports:
      - 8080:8080
    volumes:
      - ./data/demucs:/data

Best VPS for Demucs →

Related tools

Guides & articles