Pymss

Overview

Pymss is an open-source music source-separation toolchain for splitting audio into vocals, drums, bass, and other stems locally. It suits technically comfortable musicians and audio producers who want to manage models, batch jobs, and edits in one desktop workspace. Its broad workflow options come with a trade-off: Studio is still early in development and may change between versions.

Key features

Batch processing and model management

Studio can import individual audio files or folders, select a downloaded model, configure inference, and queue jobs. That makes it a stronger fit for processing collections than a tool limited to one-off uploads. Its library can browse, download, resume, and delete models, and clean up stored files—a useful set of controls when model files accumulate.

Editing and chained workflows

Waveform previews, mix controls, project persistence, and stem export let users review and adjust work within Studio rather than treating separation as the end of the process. Pymss can also chain models, with each stage using the previous output. That flexibility is valuable for users building multi-step workflows, though it asks more of them than a simple split-and-download service.

Formats and integrations

The inference package supports WAV, FLAC, MP3, M4A, AAC, Opus, Vorbis, and Ogg output. ComfyUI custom nodes and a VST3 ARA plugin extend the toolchain into other workflows. The plugin separates tracks inside ARA-capable DAWs and offers per-stem mute and solo controls; Reaper and Studio One are named as example hosts.

The WebUI is a browser interface bundled with the server and shipped with the pip package. The official site says processing runs entirely on the user's machine, an important distinction for users who want local processing. Each repository has its own license: Studio is listed under AGPL-3.0 and the inference package under MIT.

Pricing

Pymss is free and open source. The Pymss plan costs nothing and runs locally. There is no free trial because the software is already free, and there is no separate paid tier to unlock.

That removes subscription costs and upload quotas from the stated plan, but local processing means the work runs on the user's own machine. Users who prefer a hosted workflow with a clearly bounded free allowance may find a service easier to assess. Pymss's support points users to GitHub documentation and an issue tracker; Studio's README also names a QQ group for Chinese beta testing, bug reports, and discussion.

Platforms

Pymss lists Android, iOS, Linux, macOS, Web, and Windows, as well as API access. Studio is a desktop workspace, while the WebUI is bundled with the server; the project also offers its ComfyUI nodes and ARA plugin integrations.

Who it's for

Pymss is best for musicians, producers, and technically inclined users who want local stem separation, batch queues, model control, or DAW-adjacent editing. It is less suitable for anyone who needs a stable interface between versions: Studio's README warns that the app is at an early feature-development stage and breaking changes may occur.

Pros and cons

• Pro: Local processing keeps the separation workflow on the user's machine.
• Pro: Batch queues, model-library controls, and project persistence cover more of the workflow than separation alone.
• Pro: Chained models, multiple output formats, ComfyUI nodes, and an ARA plugin offer several ways to extend use beyond Studio.
• Con: Studio's early development status and possible breaking changes make it a riskier choice for users who depend on a consistent desktop workflow.
• Con: Pymss spans several interfaces and integrations; users seeking a simple hosted upload-and-download service may prefer a more direct option.

Alternatives

AI Stem Splitters and Stem Separation Software are useful starting points for comparing tools in the category.

• LALAL.AI is another free-plan option; its Starter plan offers 10 minutes in the Relaxed Queue, a 200 MB per-file upload limit, and previews, but not full downloads.
• Moises may suit users who want mobile and extension support; its free plan allows up to five uploads per month, files up to five minutes, and limited stem-separation options.
• Fadr is worth considering for a browser-oriented remix workflow: its free Basic plan includes vocals, melodies, drums, and bass, MP3 downloads, MIDI detection, and Remix Maker.
• MVSEP offers a free unregistered tier with 100+ AI models, but caps files at 10 minutes and 100 MB, limits users to 50 daily separations and one concurrent job, and requires captcha with low queue priority.
• BandLab Splitter is a simpler mobile and web alternative with four stems, unlimited track uploads, and a 15-minute file limit on its free plan.
• Suno is a mobile option whose free plan includes 50 credits per day and audio uploads up to eight minutes.
• Ultimate Vocal Remover is another free, open-source choice, with Linux, macOS, and Windows support.
• InSplitter is a web-only alternative with a free plan that caps files at 50 MB and uses standard stem quality.

Verdict

Choose Pymss if you want free, local stem separation with batch processing, model management, and tools for editing or extending the workflow. Look elsewhere if you need a stable interface across releases or prefer a straightforward hosted service; Studio's early development status is the main reason to hesitate.