A new major version of MusicBrainz Picard is here, and it brings significant changes. Picard 3.0 is live now, and the main update is more than a simple version number increase — the plugin system has been rebuilt from the ground up, alongside a major shift to Qt6 and numerous small improvements to the interface.
Today the Picard team made its announcement, and the project’s GitHub page displays the size of the effort involved. It is a program built atop the MusicBrainz database, designed to assist users with organizing and cleaning up their music collections. The latest release brings about a full reworking of how plugins are handled, along with a dark mode, improved column personalization, and fresh methods for managing cover art.
Qt6 and the Dark Mode Switch
Qt is the software framework Picard is built on. It makes cross-platform desktop apps simpler by hiding the differences between operating systems. Picard 2 used Qt5; Picard 3 moves to Qt6. That move brings real advantages, the team says.
The way the display scales has improved, so fonts no longer stretch oddly on Windows. The app now includes a dark mode that works across all operating systems, and changing between light and dark modes no longer forces a restart. This change saves time for anyone who switches modes during a single sitting.
The update makes Picard work well on today’s desktop computers. It solves the old issue where fonts stretched differently on Windows. A dark mode is now available across all operating systems, and users can switch between light and dark mode without having to restart the program.
Custom Columns and Filtering
For a long time, Picard allowed people to tailor the display of file and album lists by selecting which columns they wanted to see. Earlier editions were confined to just a handful of set columns. The arrival of complete custom column support, added by new contributor knguyen1, put an end to that restriction.
The ability to create personal columns has arrived, with scripting used to supply the actual content for those columns. A new pre-defined “Match” column is included, allowing sorting of releases by match quality. The feature’s details appear in the Picard documentation’s chapter on Custom Columns.
A fresh addition, brought in by jpmsousa03, lets users narrow down the list views. People can choose which fields to look through and supply a search term. The system then displays only the items that match the filter criteria.
Now the metadata view displays the real differences between the current tags and the new ones being written, which makes it easier to catch issues before they take effect.
Copying Tags and the Setup Wizard
The addition made by StevilKnevil now allows users to copy several tags at once from the metadata view and paste them onto a single tag, which updates all matching tags. Multiple copied values can also be pasted directly into a spreadsheet application.
New users starting Picard for the first time see a setup wizard that walks them through key configuration steps. When a feature is used for the first time, some essential functionality, including file loading and clustering, is explained by a tutorial mode. The wizard and tutorial can both be turned off or back on again from the user interface options.
The New Plugin System
The plugin system is the most significant change in Picard 3. The old system worked, but it had real limits. Each installed plugin loaded and ran code on every start, whether the plugin was enabled or not. That could lead to side effects and unexpected behavior.
The work was difficult for outside helpers too. Every piece of code needed to be placed in the main picard-plugins Git repository. The new setup splits plugin data from code, making MANIFEST.toml the required file.
The new system includes:
- Separation of plugin metadata and code, with MANIFEST.toml as the required file
- A defined plugin API with typing support
- Git-based versioning and updating of plugins
- Installation and updates from a central plugin registry, third-party Git repositories, or a local directory
- Several trust levels (official, trusted, community, untrusted)
- Central blacklisting of plugins to respond to security issues
- A picard-cli command line tool to manage plugins
- Support for a plugin translation system, optionally manageable via Weblate
- Each plugin gets its own configuration namespace
All the extension points known from Picard 2 — metadata processors, context menu actions, cover art providers, file formats, option pages — are still available. There are several new extension points too. Plugins can now register cover art filters and processors, custom variables, actions in a global “Plugin Tools” menu, and CD ripping log file formats.
In the Options → Plugins menu, installed plugins are kept apart from those still available for installation. Search works on both lists, and outside developers’ plugins may also be installed. The Plugins Options documentation explains the full setup.
This means Picard v2 plugins are not compatible and need to be migrated to the new system. There are already 30 plugins available, both migrated from v2 and brand new. The plugins v3 list on the Picard website shows the full set.
The Picard User Guide’s Plugin overview and the detailed Picard Plugin v3 Documentation both provide guidance for developers who wish to update existing plugins to Picard 3 or create a new one.
Cover Art Processing and GSoC
Users can now filter and process cover art after it has been loaded from cover art providers. The new Cover Art / Processing settings allow ignoring images below a certain size, automatically resizing loaded images, or converting them to a specific file format.
Plugins extend cover art processing by registering their own filters and processors, with filters deciding what cover art gets loaded and processors performing image processing on the cover art that is actually loaded.
twodoorcoupe built this feature during Google Summer of Code 2024, and more information can be found in their blog post titled GSoC 2024: Picard image processing.
What Comes Next
The Picard 3.0 release represents a significant change to the project. Its largest technical alteration comes from the plugin system rewrite, which delivers real advantages for both users and developers alike. The combination of manifest-based design, git versioning, and trust-level filtering provides Picard with a foundation capable of expanding while preserving established workflows.
Screen scaling and dark mode support get better with the Qt6 upgrade. Custom column and filtering features now let users control what they see. Cover art processing tools bring in automation that wasn’t there before.
| Feature | What Changed |
|---|---|
| Qt | Upgraded from Qt5 to Qt6 |
| Plugin system | Rewritten, with manifest files and a defined API |
| UI | Dark mode, consistent font scaling, custom columns |
| Cover art | Filtering and processing, extendable by plugins |
Alongside the release, a setup wizard and a tutorial mode are included to assist new users. These additions are meant to make it easier to get started with a tool that deals with a music collection’s metadata.
The release succeeds on two levels at once: the plugin system rewrite represents a genuine technical advance, while the UI changes ease the daily routine. Picard 3.0 is a real improvement, not just a numbers game.
Source material: “Picard 3.0,” metabrainz.org.
Get the Notebook.
The day's best stories and every fresh verdict, in plain English, in your inbox by seven. One email a day, no more.