Here's a snippet of the @OsaurusAI iOS app that I'm currently working on. It's being built entirely in UIKit with no traces of SwiftUI for maximum performance and backward compatibility.
Excited to bring the Mac app to mobile with Day 0 support for the new iPhone Duo!
@RyanJ_KOL@OsaurusAI The scrolling performance in the chat has been carefully crafted to handle long conversations. You should be able to scroll without any hitches even when a response is actively streaming :)
You should be able to feel the same performance on all iPhones (and iPads)!
Your AI agent isn't yours if it needs someone else's server.
In 2019, I read an essay about local-first software that I never got over. Seven years later, I built Osaurus around it.
Here's why the thing that knows how you think should belong to you.
@fromjrtosr imo SwiftUI was never anything more than Apple’s ruse to create the illusion that you needed to upgrade your iPhone for your apps to support the new features
Opened Instagram this morning. 10 ads in 7 minutes.
Every one of them wants to buy my company's data to train AI.
$100K. $1M. $2M. $4M. $5M.
Not a single one says who's buying.
@tundsdev I also don't understand why they chose to show the simulators with a solid color window now. DeviceHub should have the same windowless look as iPhone mirroring
@fromjrtosr@dhh A vertical feed containing several AVPlayer instances with auto scroll would basically be my litmus test to let SwiftUI know who's the OG
SwiftUI is monumentally bad.
One of the first things that I did for @OsaurusAI was to refactor every core feature to AppKit. The performance has been remarkable ever since
SwiftUI is not good.
It's been 8 years. We all need to admit that it was ok when we were coding, and it's just plain bad now that agents are doing the work.
UIKit/AppKit are simply better across the board in the agentic era.
Your Mac just built a game without asking the internet.
One prompt. A plan. Three files. Playable Tetris with a ghost piece and hold queue.
Pure Swift. No Python. Nothing left the machine.
MiniCPM-V 4.7 35B-A3B, fully offline in Osaurus.
This is why we're building @OsaurusAI
You own your data.
You own your inference (choose between local/cloud)
You don't have to worry about someone training on your data, even if they claim they aren't.
With local AI, your data stays on your device.