Awesome! Looking forward to the final version.

Thanks for the advice! It helped. The plugin works

That’s something I like about Smart EQ4. Choice over which track is more important.

To me, the best option was always to use TDR Nova.
You have 4 (commercial version = 6) frequencies with full control over Q, and all the compression parameters per band.
These will be the frequencies in the vocal or solo track that should be protected, find them out, and also those in the instrumental track, that are very dense and crowded. 4-6 should be enough for these perhaps 1-3 dB of impact when “demasking”.
Anything further, suggests to work more on the track separation during the conservative mixing action.
Autopilot mode gives you hardly an artistic choice.
One needs a concept about how the vocals and the instrumental should interact at all. Which frequencies on what side should typically prevail, etc. and it depends on the genre.

In FL Studio click on the VST Wrapper settings (click the cog at the top left of the plugin if you don’t see the option), go to the troubleshooting page and turn on “fixed sized buffers”.

But I found that it adds a bit of coloration (distortion) especially on the lower end. (I’m not sure if it is only on my system)