Kagi is ending development on Orion for Linux and Windows :(

Very unfortunate news: was looking forward to another browser based on webkit available for the Linux community. The good news is that they’re at least open sourcing the code for the community.

TLDR; Kagi is ending development of Orion for Linux and Windows, but they’re open-sourcing both so the community can carry them forward. Kagi is stepping back from developing these platforms directly so the team can put more focus on Orion for macOS and iOS.

<Full Article in this spoiler>

Today we have an important update. We are ending development of Orion for Linux and Windows, and we’re open-sourcing both so the community can carry them forward. Kagi is stepping back from developing these platforms directly so our team can put more focus on Orion for macOS and iOS. Orion is built by a very small team, just a handful of developers, funded entirely by Kagi users. We also chose to build it the hard way by not forking Chromium. That independence matters more than ever. The web needs more than one engine, and more than one company deciding what a browser should be. When we announced Orion for Linux, the response was overwhelming. Many of you were thrilled, tested early builds, filed bug reports, and told us how much you wanted a WebKit-based browser on Linux that respects your privacy. Thank you. That work will soon belong to everyone. Why we’re doing this When we decided to take Orion beyond the Apple ecosystem, we knew it was a big bet for a small, independent company. We made it because we believe owning the whole experience beginning with the browser, matters. Building a cross-platform browser is something usually done by companies backed by venture capital or by large open-source communities. Kagi is neither. One thing we hear often from members is that they want us focused. They’re right. Rather than spread a small team across three platforms, we’re putting our energy where we can do our best work and opening the rest to people who want to take it further. What this means for Orion on macOS and iOS Orion for macOS and iOS isn’t changing, except to get better. The team and resources that were split across platforms are now focused on making Orion on Apple devices faster, more stable, and more capable. Expect steady improvements in the coming months. Download Orion for macOS Download Orion for iOS What open-sourcing looks like We’ll release the source code for Orion for Linux and Orion for Windows. A lot of good work went into it, and it deserves a future even if we aren’t the ones who finish it now. We’re thinking about this carefully and will release details in the next 30 days. We’ve also reached out to open-source foundations and organizations about long-term stewardship. To be clear about our role: Kagi won’t be the core maintainer. These projects need people who want to own them. Get involved If you’re a developer, maintainer, or organization interested in stewarding Orion for Linux or Windows, we want to hear from you. Reach us at support@kagi.com. What this means for current users Linux: The current Linux Beta will keep working, but it will stop receiving updates from Kagi after October 2, 2026. We don’t recommend using it as your main browser. Windows: We had planned to launch Orion for Windows in late 2026. That release won’t come from Kagi. Instead, the code will be available for the community to build on. If you signed up for the Windows newsletter, this will be the last update you receive from that list. Thank you To everyone who tested, reported bugs, and believed in an independent browser beyond the Apple ecosystem: thank you. You helped build something worth sharing, and we’re glad to share it with the community. Thank you for your patience, your enthusiasm, and your trust. — the Kagi team

blog.kagi.com/update-orion-linux-windows
enPage

Replying to @⁨Pratai@piefed.ca⁩

It’s free though? Kagi the search engine is paid, Orion the browser is free. There’s still no paid browser that I’m aware of.

The problem with YouTube isn’t that it has ads, it’s that it ran at a loss for well over a decade and THEN started putting ads everywhere, increasing the amount every few months. If your service is only profitable with a ton of ads, put those ads there right away. The enshittification once you have a captive audience through the network effect is what sucks. We could have competitors and that would improve the quality for everyone, but Google sank billions into making sure nobody is using anything but YouTube and THEN started bombarding us with ads. That’s quite literally billions of dollars spent on making the market worse.

Replying to @⁨neversaydie@programming.dev⁩

they are spread way too thin

So they focus on macOS, the least used platform?

I’m not in favor of Windows exclusive applications but if you only have the resources to focus on one platform, use the one with most users.

As for the Linux version: Looked like yet another Gnome Web knockoff. A WebKit browser whose GUI actually works well on Plasma and other non-Gnome desktops would be welcome but another one of those isn’t a benefit IMO

Replying to @⁨Sunny@slrpnk.net⁩

I read the article. They’re open sourcing the Windows and Linux versions, so the community can take over if they want to. So it will live or die based on what the community actually wants from it.

As a Mac user, it’s weird to see a company drop the other two in favour of maintaining the Mac version. I don’t use it, or Kagi, but I’m aware of both. It’s a good platform but we don’t have the users. Same with Linux, but it runs on more hardware, and bleeding the Windows market share benefits everyone, including Windows users.

As for iOS, there’s no winning the browser war there. It’s either Safari or compromise. I like Firefox, but it doesn’t/can’t do enough. DDG is good for YouTube because it blocks the ads. But mostly I just use Safari. On my Android phone, I use Firefox with uBlock Origin, naturally.

Replying to an earlier post

Honestly it’s not as bad as people make it out to be and it’s also worse all at the same time. The gestures are awful, the animations take too long even if you reduce them and the keyboard has autocorrupt with the added bonus of not registering input properly. The three main positives for me are: the battery (on standby time), video/camera are very good, and the shortcuts app which enables you to create things that are missing in iOS.

Oh and certain apps do get preferential treatment on iOS but the open source space is quite barren compared to Android.

Replying to @⁨fatcat@discuss.tchncs.de⁩

Ye, you can only install 3 apps without exploits tho (I used to use lara until i broke my phone doing something very dumb and had to update it(the hardware broke not the software) it does only work for like a bit of 17 to a bit of 26 or something, before that you can probably jailbreak anyway) but you can use livestore and livecontainer as 1 app, livecontainer lets you install multiple apps in 1 container, basically bypassing the limit, problem is they’re on the same sandbox so be careful with personal info and that. It also doesn’t work for all apps.

You can also make shortcuts to apps inside livecontainer(s). Also you do need a pc or laptop for the initial setup but it’s actually mostly incredibly easy unless you’re the kind of people who don’t know to install a .exe file (I’ve seen people do that(i don’t even main windows))

Also idk if apple is trying to break this stuff but I’ve heard some stuff breaks sometimes, I didn’t experience it cuz I didn’t actually update my devices but yeah.

You normally can’t use JIT but you can activate it but it does use the vpn slot and you need to sideload something like stickdebug I think, idk if that’s what people are still using nowindays, it also takes the vpn slot to install apps. you need a specific vpn app that’s on the app store.

Idk if this was very useful, specially after 13h but yea.

Replying to @⁨iopq@lemmy.world⁩

I suspect we’re on the same side here with only a slightly different perspective. I’m on complete agreement that ladybird, aside from its controversy, and servo represent way more diverse entries for new browsers to consider. I’m just saying WebKit is better for diversity than blink. Only by 5x or so but still, that’s something. Making the decision to start a browser today, 100% I’m going with servo.

Replying to an earlier post

WebKit is not starting from scratch. It goes back to KHTML which was started by KDE for their web browser back in the 90s. Apple built Safari on top of KHTML, forking the engine and creating @webkit@front-end.social. WebKit continues as an open source project, with native web integration in GNOME running through @WebKitGTK@floss.social. Ironically KDE and Konqueror stopped using WebKit, using Chromium these days. But WebKit is, at least to a degree, a natural choice for anyone interested in making an independent browser for GNOME or Mac OS.

Blink, the engine behind Chromium, is forked from WebKit, but it’s monopolizing the internet and is generally worth avoiding. I don’t think anyone is able to separate Gecko from Firefox, so a Gecko-based browser would just be a Firefox fork. Meanwhile @servo@floss.social competes directly with the KHTML family of browser engines by being more flexible, but it’s not ready for serious use yet. Once Servo is ready I suspect we will se a lot of independent browsers popping up.

Contributing to Firefox doesn’t make much sense these days. Mozilla, or at least Firefox, runs their own ship and have turned their back on the open source community. Those concerned about the open web are getting increasingly worried that Mozilla is not going to do a whole lot to counter Google’s influence, leaving it to Google to take the internet and destroy it in any way they see fit. If Mozilla does not fight for us any more the only thing we can do is to double down on independent engines such as WebKit and Servo.

Replying to an earlier post

building a browser from scratch is just an insanely difficult task to begin with. Building a browser from scratch tailored to each specific platform instead of doing something like QT is just idiotic. There’s a very good reason why the majority of todays browsers are forks of Chromium or Firefox and why very few are their own from scratch. Building a web browser is probably one of, if not the most, complicated bits of software to build other than building an OS from scratch.

I’m not a fan of Kagi, I can’t justify paying for a search engine especially after honestly trying it and getting mediocre search results. I was optimistic about Orion but after following the development for awhile I knew it wasn’t going to pan out. they spread themselves too thin on that project and it sows further doubt for Kagi overall.

Replying to @⁨victorz@lemmy.world⁩

If I’m not wrong, it’s not complicated. But maybe I’m missing something. I don’t think I can easily see in which way you are looking at it in your example. 🫣

I’m looking at it from both ways, because as I said, it’s complicated. Half of the companies you deal with these days are using AI in some form, whether it’s your internet provider, TV manufacturer, local grocery store, etc.

As another example: Your local grocery store used AI to write short descriptions for products for an ad or display. First of all, can you even tell it’s AI from a short description…? And second of all, does that mean buying groceries from that business is funding AI…?

I personally don’t think having a Kagi subscription is considered “funding AI”, especially considering another comment that points out they are transparent about where your subscription goes, but depending on where you draw the line, maybe you would.

Replying to @⁨victorz@lemmy.world⁩

It means any dollar works the same as any other. That’s how I would put it.

Here’s a dictionary:

www.merriam-webster.com/dictionary/fungible

So if Kagi takes a bunch of dollars in and makes a claim that THIS particular dollar funds the AI project, but THAT particular dollar does not fund it, that claim about the specificity of the individual dollars is on its face bullshit.

Replying to @⁨victorz@lemmy.world⁩

That would only be if they charged you more on top of your subscription for the AI usage. If they send part of your sub to the AI company, and don’t reduce your usage of their non AI features, then really it’s just other people not using AI who are subsidizing the chunk kagi lost to the AI company.

If it were truly fungible they wouldn’t be providing subscriptions and it would all be usage based. Subscriptions just allow the users that don’t search as often to subsidize the power users, making Kagi come out ahead overall.

Replying to @⁨victorz@lemmy.world⁩

I guess to be fair, Kagi does have two different AI components. They have their Kagi Assistant which is your bog standard AI chat interface, which does cost extra and is displayed in the details page you are talking about.

But they also have Quick Answer which is like the google AI overview that appears when you make a search. It will summarize the results for you. It auto triggers when you end your search with a ? or if you click the Quick Answer button. It can be disabled. However it does not get charged extra for, it is part of your subscription.

So its kind of both.

Replying to @⁨shifty@leminal.space⁩

As someone in the same position, it’s a difficult line to manage.

I’ve criticized Kagi before, but it’s still unfortunately the best model for search on the market. I pay them money, they give me search results. As the prior post said, they’ve introduced switches to disable all of the AI. And if you believe what they’ve said about costs, then search is more expensive for them than AI — so if you use a lot of search (as I do, ≈1.2K/Mo), and don’t rack up AI usage, then that money isn’t going to AI, viewed in a vacuum of course.

Replying to @⁨Sunny@slrpnk.net⁩

Ladybird if anyone wants to watch this space, I’m most excited to see them succeed.

ladybird.org

Ladybird is a new browser engine built from scratch. Not a fork. No user monetization. Backed by a 501©(3) non-profit. Status: In active development Target: Alpha 2026 (Linux & macOS) Get involved Support us About

The web is the most important platform in computing, yet only a small number of browser engines power the entire thing, each controlled by a company with its own agenda.

Ladybird is building something new: an independent engine, driven by web standards, and free from conflicts of interest.

Our first Alpha release for Linux and macOS is coming in 2026.

What makes Ladybird different Truly independent

We’re building a new browser engine from scratch, based on web standards. Not a fork of an existing engine. Singular focus

We are focused on one thing: building a web browser. No monetization

No “default search deals”, crypto tokens, or other forms of user monetization.

Replying to @⁨esc@piefed.social⁩

It’s like Safari in terms of performance and battery usage, with extension support that isn’t pants-on-head stupid like Safari’s, better customization, and some advanced features. That’s all I really want.

Safari is just too basic for me to use as an everyday browser (especially on iOS).

I think the original impetus was that Safari makes it unreasonably difficult to change its search engines.

Replying to an earlier post

I’ve started using Gnome Web as my main browser, and I’m super happy with it. I’ve been trying it occasionally for years, and only the last month or so does it feel mature enough to replace firefox.

I guess I have the advantage that I mostly avoid bloated websites, but still. For anyone looking to escape the web engine duopoly Gnome Web is very much worth checking out these days.

Replying to @⁨Sunny@slrpnk.net⁩

I think this is the right move. They started on MacOS. They expanded to Linux and Win for reasons that were unclear (to me). If they can’t keep all three going, open-sourcing the code is appropriate. If people want to keep going, awesome. If not, well… And they refocus their resources to the platform that they presumably consider their primary home, else they wouldn’t have started with MacOS first.

I have Orion installed and occasionally use it. But I’m also a browser junky and can’t resist trying every new browser I read about. (I’m posting this from the recently announced Search browser and it’s using impressively low system resources. I recommend test driving it if you’re curious. I’m not affiliated in any way).

Replying to @⁨Sunny@slrpnk.net⁩

Personally, I just don’t see any point in developing anything for closed platforms like MacOS or iOS.

At the end of the day, you’re just making Apple more money for free, on very bad terms. They have all the control; you’re just an unpaid intern.

Who knows when Apple will stop supporting half the libraries your project relies on?

Instead, you can develop for open platforms, which genuinely makes the world a slightly better place. In addition to you having a guarantee the project won’t have to shut down overnight.