https://fluxer.app/ -- this project is ticking all the boxes for me for a discord replacement

- explicitly designed with discord-style use cases in mind (community/guild oriented with multiple chatrooms per community, voice channels + screensharing as core features) -- it pretty much just clones discord's UI, but this makes it very familiar for discord users, which is a plus for adoption

- fully open source (AGPL, even!) and self-hostable, to be enabled by...

- federation under development (roadmap has a bit more info https://blog.fluxer.app/roadmap-2026/) -- this is important to make self hosting an actually viable and attractive option

- main instance not hosted in a US jurisdiction

- most importantly: almost all of the core features _already work_ as of today. screensharing in voice channels has some limitations but the main dev has been saying they'll be rolling out updates soon, and being open source means it's possible for motivated devs to contribute

here are the main drawbacks it currently has compared to other solutions:

- the main instance is a bit unstable at the moment because they didn't anticipate the influx due to discord's announcement. availability hasn't been great over the last 2 days. i expect this will improve as they get on top of the increased load and as contributions to their paid tier start coming in

- not e2e encrypted. i'm fine with this, but you might not be. however they do have what they're calling "secret channels" on the roadmap which will be e2ee.

- no native mobile apps at the moment, just PWAs (you can go to the app on your phone's web browser and then save it to your home screen and it basically works like a normal app)

- serving users in age verification required jurisdictions means that, at some point, they will likely be legally required to do age verification. however, i think self hostability makes this much less of a liability than it would be on a closed source service

- currently, it seems to be a single dev project. it's wildly impressive what they've managed to accomplish but the dev getting overloaded, burnt out, or changing the project direction are potentially liabilities. again though, the app being open source provides a potential mitigation)