Picard not responding and pc runs out of memory on a single track

I am trying to tag a single title and Picard crashes with out of memory error.

Steps to reproduce:

  1. Open Picard
  2. Browse to Recording “I’m Alright” by Kenny Loggins - MusicBrainz
  3. Click tagger.
  4. Wait for the crash to happen! *1

*1 after a few minutes the [loading album information] shows, I see the memory in windows task manager increasing until there is no more memory left on my PC (8GB) and everything frozes.

It happens only with this track. Any other track works well.

Can someone try it please?

I think I have found where the problem was:
I had enabled “User release relationships” and “Use track relationships” in Metadata options.
I disabled these and now it does not crash.

Thanks for reporting, but I cannot reproduce this. A few questions:

  1. Do you hit the tagger button on the release or the recording?
  2. Which plugins do you have enabled?
  3. What is you tagger script?
  4. Which version of Picard and what operating system?

Thank you for taking the time to reply to my message.
I hit the tagger on Tag Lookup Results - MusicBrainz the release “The Radio Essentials Series”.
It’s a very large collection with 60 CDs.

I had enabled the following plugins:

AcousticBrainz Tags
Deezer Cover art
TheAudioDB cover art
Wikidata Genre

I don’t have any tagger script.
The versions are: Picard 2.7.3 , OS windows 11 21H2 build 22000.593

I have now enabled only AcousticBrainz Tags Plugin and enabled again “Use release relationships” and “Use track relationships”.
My mistake was that I hit the release tagger button and not the recording.
I’ve been using Picard only for three days. Still learning.


Thanks for the details. The most likely cause for this is the wikidata genres plugin. Depending on the plugin’s settings it will try to load the genre for every artist, every linked work and the release group itself. Each such lookup requires one call to the MusicBrainz web service, and one call to Wikidata. The MusicBrainz web service is rate limited at one request per second. So for a release like this it will take a significant time to finish.

As for the memory grow I’m not sure. The plugin caches genre lookup results, but that should not grow into giga bytes even for this release. Might be some issue that can be solved, either with the plugin or with Picard. I’ll try to reproduce.