@diego I'd like to continue talking about the possibility of using RRML for Automoderator. When we discussed this with @isarantopoulos and @Pablo a few months ago it sounded like there might be some performance concerns, but I'd like to get some clarity on those.

What are the current performance concerns with RRML, is it only the length of time for a response? Since we're operating asynchronously post-edit, this isn't inherently a problem for us - it doesn't matter if it takes a few seconds for us to get a response. I guess the only concern would be if we overload LiftWing in some way. Do you have any additional info/guidance on how we can evaluate whether we can/should use RRML?

Hi @Samwalton9-WMF , we choose RRLA because it was more stable, but since then, we made some updates to RRML (it was not only about serving time, but getting errors for some revisions), that aimed to make it more stable.
So, if there is interest to switch to RRML (for 47 languages with coverage), my recommendation would be to run some stress test on that service, and measure the % of errors and if Automoderator can tolerate them.

@Samwalton9-WMF , just keep in mind that the scores from RRML and RRLA are different. This means that you maybe need to run new users' test to (re)define the thresholds.

This is just waiting on T400727. Folks from es.wiki and uk.wiki have both expressed interest in switching to the new model due to community perceptions of the language agnostic model.

Awsome news, looking forward to the moment we can use Automoderator on rowiki as well.