You can run DeepSeek Flash just fine on 128GB using lower quants. Antirez' DwarfStar supports it really well. The bigger the model, it seems, the less it is affected by high quantization. I know some people are working on quantizing GLM 5.2 so it can fit into 128GB (using additional tricks to select only some experts, etc). There's a lot going on.
I'm using this in one of my side projects, Emyn ( https://github.com/terhechte/Emyn ) a macOS virtual camera app for composing camera video, app windows, backgrounds, effects, notes, and captions into a polished live presentation feed.
It works very well, the integration is much easier than before, users have model choice. So happy that this exists!
I'd rather have a ai-written article then no article at all. I think this is a real struggle. Writing an article takes time that could be spent on improving the product. Would I prefer a human article, yes. But if the alternative is "no article", then that's how it is.
How is Peter "early in their career"? When he sold PSPDFKit for 100mio in 2020 he had been working on it for 13 years, and before that he'd worked as an engineer.
Instead of win key, you can press F3, or just set a hotkey that works for you in the System Preferences
Instead of clicking the red maximize button, you can double-click the window header / title. This will use an algorithm to try to resize the window to the best size for its content.
I like doing side projects, I don't like wasting a day of work potential on any of these web apps: Google Cloud, AWS, Azure, Appstore Connect, Google's Android App Store, RevenueCat, Stripe, etc
I dread having to log in to these systems and waste hours achieving the simplest tasks.
This is what I'm using Claude for. E.g. I log in to AppStore connect, tell it what I need (3 subscription tiers), it will do all the clicking and editing and Apple's stupid UI, then I will ask it to create a summary for RevenueCat, and use another Claude session in there to click all the buttons to configure based on what just happened in Appstore connect.
I tried a virtual boy the first time some months ago. I was super impressed with the quality of the graphics. I've used a Quest 3 and a Vision Pro and obviously these are very different beasts, but its just impressive that in terms of sharpness, this 90s device felt as good.
KDE has a setting to switch the cmd & command keys so that e.g. command+c copies instead of ctrl+c. This works in all KDE apps (it will not work if you install any Gnome/GTK app, though). I forgot the setting but its something in advanced and used to be called Emacs key binds, but now I think it just refers to the keys.
Anyways, beyond that, have a look at Kinto which tries to do everything in one box, but it is an additional software you have to run:
There're multiple task solutions for Claude or other llms that let it define tasks, add implementation notes and (crucially) add sub-tasks and dependencies. I'm using Beads (https://github.com/steveyegge/beads) and I think it really improves the outcome; especially for larger projects.
There's a mix of both worlds that I've tried for a while and want to pick up again in 2026:
Use macOS so that I can utilize the great hardware and the well integrated drivers (e.g. sleep, performance, silence), but then for each project / work stream just fire up a lightweight linux VM fullscreen and do everything related there. E.g. all browser windows/tabs, apps, file explorer windows, terminal sessions.
When I stop working I pause the VM. When I need to continue everything is as I left it.
The main reason why I stopped was that the 2d hardware acceleration for Linux didn't work in UTM.app. I think I'll just need switch to Parallels or VMWare
Mostly Swift, Rust nowadays.
Podcast: https://contravariance.rocks
Twitter: https://twitter.com/terhechte
Mastodon: https://mastodon.social/@terhechte
[ my public key: https://keybase.io/terhechte; my proof: https://keybase.io/terhechte/sigs/ETCgR5hPJ0vXdK9SGFqWuMxn_RglDOVJbeJbseoD-4s ]