Plans for every workflow

Start free. Upgrade when you need more. No hidden fees.

Free

$0/month

Try it out. No credit card required.

  • 3 separations per day
  • MP3 output
  • MDX-Net Q model
  • Max 5 min track length
  • Quick Voice Lift & Denoiser
Get Started Free
MOST POPULAR

Basic

$5/month

Great for hobbyists who need more length and quality.

  • 30 separations per month
  • WAV output
  • Demucs v3 & HTDemucs models
  • Max 10 min track length
  • HTDemucs Voice isolation
  • Standard processing queue

Pro

$18/month

For producers & DJs who need high volume & all models.

  • 100 separations per month
  • WAV / FLAC / MP3 output
  • All models incl. 6-stem & Hybrid
  • Max 30 min track length
  • Fine-tuned Voice & DNS64 Denoiser
  • Priority processing queue

Frequently asked questions

What audio formats are supported?

We support MP3, WAV, FLAC, OGG, AIFF, and M4A. Files up to 100MB are accepted on the Free plan.

How does the AI model work?

We use Meta's HTDemucs — a Hybrid Transformer architecture trained on thousands of hours of multitrack audio data. It's one of the highest-quality open source separation models available.

Can I use the stems commercially?

The stems you generate are derivatives of the original song. Commercial usage rights depend on the original track's license. We do not impose any additional restrictions.

Is my audio stored on your servers?

Audio files are processed in-memory and automatically deleted within 1 hour after separation is complete. We do not retain or analyze your files.