Psychotoxical
b6ac879b14
* feat(settings): add SettingsSubCard primitives for sub-section controls Reusable wrappers (SettingsSubCard / SettingsField / SettingsRow / SettingsValue / SettingsCallout) that encapsulate the settings-norm-* sub-card styles. Per-mode and detail controls inside a settings section should use these instead of hand-rolling the box with inline padding or one-off classes — the lack of a shared primitive is what let recent additions drift off the pattern. * refactor(settings): move the Audio tab onto the SettingsSubCard primitives Normalization, Track transitions and Hi-Res now build their sub-cards from the shared primitives instead of raw settings-norm-* markup. Track Previews, which rendered its sub-options bare with inline styles and dividers, now uses the same sub-card so the whole Audio tab is consistent. Behaviour and styling unchanged; Track Previews' start/duration become labelled slider rows like Normalization. * refactor(settings): move the remaining tabs onto SettingsSubCard Apply the shared sub-card primitives across the rest of Settings so every tab matches the Audio-tab reference look: - Appearance: grid columns, window buttons, UI scale, font, seekbar - Library: random-mix blacklist, rating-filter detail - Input: in-app + global keybind lists - Integrations: Discord templates, external-artwork BYOK - System: language, Linux Wayland text render, clock format, logging - Storage: media dir, hot-cache detail, downloads folder - Backup: mode picker + per-mode export/import Detail/per-mode controls now sit in the bordered sub-card; toggle groups stay on SettingsToggle. Behaviour unchanged. * refactor(settings): box the data-heavy and shared sections in SettingsSubCard Bring the previously-skipped 'borderline' sections onto the sub-card too: - Playback speed: wrap the settings-mode controls in a sub-card; the compact player-popup path is left untouched (shares the component via the compact flag) - Theme scheduler: mode picker + theme/time selects move into the sub-card - Cover-cache + analytics strategy: the per-server tables sit in a sub-card Behaviour unchanged; the table layouts and the playback-rate-* internals stay as they are, only the surrounding sub-card is added for visual consistency.
A modern desktop client for self-hosted music libraries
Fast. Native. Beautiful. Built for people who actually care about their music collection.
Psysonic is built primarily for Navidrome and also works with Gonic, Airsonic, LMS and other Subsonic-compatible servers, depending on the features supported by your server.
Available languages: English, German, Spanish, French, Norwegian Bokmål, Dutch, Romanian, Russian, Chinese, Japanese and Hungarian.
More translations are added over time.
No telemetry • Native performance • Navidrome-first • Community driven
What is Psysonic?
Psysonic is a desktop music client for self-hosted music libraries. It is designed for people who want the freedom of their own server without giving up the comfort, polish and speed of a modern music app.
It is built with Rust, Tauri v2 and React, with a strong focus on responsiveness, customization, practical music-library workflows and a user interface that does not require a manual before you can press play.
Psysonic is optimized first and foremost for Navidrome. Other Subsonic-compatible servers can work well too, but advanced features may depend on server-side support.
Highlights
Playback & Queue
- Gapless playback
- Crossfade
- ReplayGain support
- LUFS-based Smart Loudness Normalization
- AudioMuse-AI support
- Infinite Queue
- Smart Radio sessions
- Fast and responsive playback handling
- Low memory usage compared to heavy web-first clients
Audio Tools
- 10-band Equalizer
- Equalizer presets
- AutoEQ headphone correction
- Per-device optimization
- Loudness-aware playback options
Library Management
- Fast search across large libraries
- Albums, artists, tracks and genres
- Ratings support
- Multi-select bulk actions
- Drag & drop playlist management
- Smart Playlists
- Built for large self-hosted collections
Lyrics & Discovery
- Synced lyrics with seek support
- Lyrics provider support: YouLy+, LRCLIB and NetEase
- Auto-scrolling sidebar lyrics
- Fullscreen lyric mode
- Last.fm scrobbling
- Similar artists
- Loved tracks and listening stats
Sharing & Social Listening
-
Magic Strings sharing:
- share albums, artists and queues
- Navidrome user management helpers
- fast account sharing
-
Orbit shared listening sessions:
- host-controlled synchronized playback
- session invites via link
- guest song suggestions
- real-time queue interaction
Personalization & Accessibility
- Large theme collection
- Catppuccin and Nord inspired styles
- Glassmorphism effects
- Font customization
- Zoom controls
- Keybind remapping
- Theme Scheduler for automatic day/night switching
- Colorblind-friendly theme options
- Keyboard-friendly navigation
Power User Extras
- CLI controls
- USB / portable sync
- Backup and restore settings
- In-app auto updater
- LAN / remote auto switching
Orbit brings synchronized shared listening sessions directly into Psysonic.
Start a session, invite others with a link and listen together with host-controlled playback, shared queue interaction and guest song suggestions. It is built for real-world music sharing without turning your self-hosted setup into a social-media circus.
Platforms
| OS | Support |
|---|---|
| Windows | Native installer / WinGet |
| macOS | Signed DMG |
| Linux | AppImage / DEB / RPM / AUR (psysonic, psysonic-bin) / NixOS |
Install
Linux
curl -fsSL https://raw.githubusercontent.com/Psychotoxical/psysonic/main/scripts/install.sh | sudo bash
Linux builds are also available through GitHub Releases, AUR and Cachix/Nix.
AppImage runs under X11/XWayland — it pins
GDK_BACKEND=x11for a stable WebKitGTK stack. For a native-Wayland launch, use the.deb,.rpm, AUR, or Nix packages, which follow your session's display server.
Windows
Download the latest installer from the GitHub Releases.
or,
install via Windows Package Manager (WinGet):
winget install Psysonic
You can also browse and install it on winstall.app.
macOS
Download the signed DMG from the GitHub Releases.
Development
Contributor expectations (PRs, CI, Tauri boundary, UI): CONTRIBUTING.md.
git clone https://github.com/Psychotoxical/psysonic.git
cd psysonic
npm install
npm run tauri:dev
Build release:
npm run tauri:build
Privacy
Psysonic is built for self-hosted music collections. Your library is yours.
- No telemetry
- No spyware nonsense
- No analytics harvesting
- No hidden tracking
See TELEMETRY.md for the telemetry stance and PRIVACY.md for how each opt-in integration handles data.
Reviews
Community & Support
Join the community, report bugs, suggest features, share themes and help shape the future of Psysonic.
License
Psysonic is licensed under the GNU GPL v3.0.
Forks and Attribution
Psysonic is free and open-source software under the GPLv3. You are welcome to fork it, modify it and build upon it under the terms of the license.
If you publish a modified or rebranded version, please make it clear that your project is based on Psysonic and preserve proper attribution to the original project.
That is not about preventing forks. Forks are part of open source. It is about being honest with users and contributors about where the work comes from.
Features, design work and implementation ideas developed in Psysonic should not be presented as unrelated original work in downstream projects.
Own your music. Enjoy the client too.
Psysonic brings a modern desktop experience to self-hosted music libraries.