Mattermost is decent, but the original intention was a replacement of Discord. At the time (around 2024), the best option for us was Mattermost for text and Mumble for VoIP. Now in 2026 the situation has greatly evolved, we have many alternatives and it's a better time than ever to self-host a fully fledged alternative. Yet, Mattermost isn't going anywhere for the short term. It's currently an established chat-solution and at the bare minimum will be fully support for atleast a year, if not more. But of course a full replacement would be ideal. The current service will transition into an active archive in that case, remaining online for historical purposes. > ⚠️ This is more like a draft at this point. The following projects seem like good choices for out use case: | Solution | GitHub | SSO Support | Mobile App | Desktop App | E2EE | VoIP | Viability | | --- | --- | --- | --- | --- | --- | --- | --- | | [DCTS](https://dcts.community/) | https://github.com/hackthedev/dcts-shipping | ❌ [User-base has seemingly no demand for it](https://github.com/hackthedev/dcts-shipping/issues/43) | ⚠️ Android-only | ✅ Windows/Linux/macOS | ✅ For DMs | ✅ Voice Channels w/ Video & Screen-sharing | 🟡 Small project, primary dev advertises in other subs | | [Haven](https://ancsemi.github.io/Haven/) | https://github.com/ancsemi/Haven | 🚧 On the [Roadmap](https://github.com/ancsemi/Haven/issues/18), no ETA | ⚠️ Android-only | ✅ Windows/Linux | ✅ For DMs | ✅ Voice as part of Channels w/ Screen-sharing | 🟡 Small project, [Extensive AI usage](https://github.com/ancsemi/Haven/issues/5440#issuecomment-5003453705) | | [Stoat](https://stoat.chat/) | https://github.com/stoatchat/self-hosted | 🚧 No clear information, [apparently WIP](https://github.com/stoatchat/stoatchat/discussions/507#discussioncomment-15710135) | ⚠️ Hardcoded server-URL | ⚠️ Hardcoded server-URL | ❌ | ✅ But no video/screen | 🔴 Complex self-hosting stack | | [Fluxer](https://fluxer.app/) | https://github.com/fluxerapp/fluxer | ✅ | 🚧 | ✅ Windows/Linux/macOS | ❌ | ✅ Voice Channels w/ Video & Screen-sharing | 🔴 Complex self-hosting stack | | [Chatto](https://chatto.run/) | https://github.com/chattocorp/chatto | ✅ | 🚧 | 🚧 | ❌ | ✅ Room Calling w/ Video & Screen-sharing | 🟠 [100% Agentic](https://hmans.dev/blog/chatto-is-robots) | ## Annotations ### Stoat (formerly Revolt) This has been a mess to self-host in the past. I am including it in here more as an element of comparison, not a realistic choice. Similar would be Spacebar, but this one doesn't qualify due to not being a production ready project, seems more like a for-fun rewrite of Discord if anything else. ## Requirements for becoming a libre.moe service I really don't want to deploy a platform that requires >= 5 different microservices because I do not know how well they will all be integrated and how well upgrades are tested. There are already a decent amount of services, which are currently well maintainable and low complexity, and I want to keep it that way. Something like Flxuer or Stoat with their >10 services (not even counting reverse proxy and db) is simply out of scope for me.