New improvements coming to Mullvad VPN. So glad to see them to continue innovating and improving themselves. Wish all tech companies behaved the same way - to have a real business model and improve or continue to provide tangible value to its users.
The impact has been immediate. So far not a single crash has stemmed from GotaTun, meaning that all our old crashes from wireguard-go are now gone. Since rolling out GotaTun on Android with version 2025.10 in the end of November we’ve seen a big drop in the metric user-perceived crash rate, from 0.40% to 0.01%, when comparing to previous releases. The feedback from users’ have also been positive, with reports of better speeds and lower battery usage.
Mullvad can’t stop winning
Glad to hear. I saw some stuff in the Proton roadmap of them doing some sort of revamp on their VPN as well.
Yeah, they need to innovate. More and more tech savvy and trustworthy YouTuber’s are mentioning Mullvad and not Proton. There’s a reason for that.
I’m excited to see what Proton has up its sleeve. It better rival Mullvad. If it does, then with its streaming and port forwarding capabilities, it will again come back on top for me in some use cases. Though I must say torrenting works just fine on Mullvad too even though they don’t have port forwarding anymore.
tbh, in our app’s crash reports as reported by Google Play, wireguard-go barely registers. This isn’t a commentary on WireGuard’s stability, but just a counter data point.
And in our app’s test env setup (on Linux not Android), wireguard-go can do 1gbps or more, just fine. On Android, we are limited by other things, not wireguard-go.
Do you have a Linux app I don’t know about? Not sure how you mean here.
We build and test our network engine viz. firestack on Linux, specifically on just Ubuntu.
Come to think of it, we may as well package and release it (but then, the CLI does not have all the features that the Android app has and we’ll have to implement those to the extent possible).
What I meant to highlight was, the Mullvad apps are already a mixture of Rust and Go and other programming languages. When you put all of this together, they might have ended up with more issues than warranted. Removing either Go or Rust would have been a good decision, but getting rid of Go given DAITA is all Rust makes sense for just that. Especially since Go is packaged up as “cgo” to be run on Android, which some purists say isn’t really Go anyway (:
In our test env, wireguard-go run as-is (no cgo / no FFI interfacing with 20 other languages), and it is pretty stable. Mullvad throwing shade at wireguard-go is … their marketing instincts taking over. Going by WireGuard creator Jason Donenfeld’s public comments and interactions with the community, they seem super serious about stability and competent enough to accomplish that goal.
Thank you for explaining.
From looking at Mullvad’s repos, they do seem to favor Rust for their in-house implementations, so trying to converge there does make sense.