Top
Best
New

Posted by allanrbo 13 hours ago

Codex in ChatGPT desktop app for Linux is now in preview(community.openai.com)
394 points | 269 comments
jtrn 1 hour ago|
So much negativity... I like the desktop app and am looking forward to testing it out.
moronicles 8 minutes ago|
Remote alone makes the desktop app worth it.
miguel-muniz 3 hours ago||
Unrelated to this Linux release but I've been using the new ChatGPT desktop app since I annoying updated on accident, and I haven't had that great of an experience.

I previously really enjoyed the Codex app. It was easy for me to understand and find what I needed. The conversation view provided the right amount of information without feeling too concise or full of jargon and the typographic hierarchy made it easy to skim. The project based workflow made sense for my needs.

Ever since they combined all ChatGPT into one app though it has become a mess. It's trying to do too much in one. The settings menu is especially bloated, and I find the UI to be too complicated for the simple questions I'm asking outside of development work. I would rather just use the web interface for general use at this point. They've also been updating much of the UI to follow the same patterns as Claude, and I've not been impressed with the Claude desktop app.

To me, they are taking multiple steps backwards. I hope they eventually find their way again.

ninkendo 4 hours ago||
Sitting at 2.2GB of combined resident RAM across 11 processes. I haven't done a single query yet.
ChadNauseam 18 minutes ago||
Is this a linux issue? My Macbook has 16gb ram and running out has literally never been an issue for me. I'm using all of it + 10gb of swap and my computer feels snappy and fast. Chrome with 100+ tabs, discord, slack, codex, chatgpt, all running simultaneously with no issues. Sure, all other things being equal less RAM usage is better, but I can understand why reducing it isn't a priority for them.
reaperducer 32 minutes ago|||
Sitting at 2.2GB of combined resident RAM across 11 processes. I haven't done a single query yet.

Surely they can just vibe is down to a reasonable footprint, right? Right?

geodel 4 hours ago||
But you might. And Codex is taking your threats seriously and stand prepared in case you breach protocol and start using your computer resources for your own nefarious purpose.

These apps stands as last defense between users and legitimate owners of compute resources that is electron and pals.

DefamationStati 16 minutes ago||
Finally! My shitbox Surface 8 pro that's running Debian will be pleased.
WalterGR 10 hours ago||
From yesterday, with 26 comments: https://news.ycombinator.com/item?id=49264334
Beijinger 2 hours ago||
Not sure what I should see under the Link provided. Here is another one:

https://community.openai.com/t/codex-in-chatgpt-desktop-app-...

"The desktop app is designed as a workspace for managing projects, working with files, using browser workflows, and running Codex alongside ChatGPT."

Not sure if I really understand this. And somehow I am worried about ChatGPT having access to my files and workflows. How much control do I have?

pdhborges 10 hours ago||
For someone who has not tried these desktop apps: what is the the advantage versus cli codex with some MPCs and multiple folders each one with their context files?
Gareth321 10 hours ago||
The same as an OS with a UI. Discoverability, fewer actions per activity, easier navigation, faster to learn/lower skill floor, etc. Some highlights:

1. Much better multi-agent orchestration. The app supervises many agents, with different project views and built-in git worktrees, so agents can work on isolated copies of the same repo. It's easy to jump between tasks and see changes. I recall OpenAI wrote that CLI was not designed around this style of work.

2. The app combines multiple terminals, file previews, PR-review workflows, SSH connections, an integrated browser, and page annotations. For front-end stuff, Codex can run the app in Codex for mockups/prototypes, then iterate inside the app. IMHO I think this is a much faster, more intuitive flow.

3. For long-running and recurring agent workflows, I think they're much easier to manage in the app. You get automations, review queues, persistent conversation context, future schedules, and auto resume for longer running tasks. E.g. CI failure summaries, issue triage, and release briefs.

You can do some of the above, to some degree, with varying levels of success in the CLI, but I think it's miles ahead in the app.

KronisLV 5 hours ago|||
> The app supervises many agents, with different project views and built-in git worktrees, so agents can work on isolated copies of the same repo. It's easy to jump between tasks and see changes.

This is mostly true with Codex and Claude Code Desktop, but gets a bit trickier with 3rd party tools.

I tried out some others and both Paseo and Kepler felt a little bit clunky especially around sub-agents and workflows (even if they let me have both OpenCode and Claude Code with different providers in the same UI). And OpenCode is plain lovely in the terminal but sorta okay in the browser: compare its display of sub-agents to what Claude Code does, for example; their tabbed interface also feels a tad less pleasant than the vertical menu of projects + sessions that most others are settling on, not their old look with icon per project, that was worse. There's definitely ample room for better tooling, then again early IDEs were far worse, so hopefully ADEs will get more polish eventually!

Still, regardless of what you use and across how many providers, it's nice to have everything in a single familiar interface, if you need that.

jtrn 1 hour ago|||
Nonono. You are supposed to pile one with negativity and point out everything that is bad with it!!
arminluschin 10 hours ago|||
You get GUI niceties: a nice presentation of the output like inline markdown rendering including tables and images in a nice font. Input supports drag & drop, and I can actually navigate the input field the way I am used to (pointing and clicking, using keyboard shortcuts to jump to beginning of words / sentences). Not sure what the most modern CLIs support, but for me just the ergonomics of a GUI text field alone are superior.
Pavilion2095 10 hours ago|||
It's easier to use for most people. I don't think CLI and GUI programs target the same audience.
logicallee 10 hours ago||
I'm not ready to try this yet, since after some recent experiences[1] I've decided not to be at the cutting edge of new tools and models for a few weeks or months.

However, from the screenshots and description, I expect that you would have these advantages:

- History. Currently, if you type into the CLI (TUI) your intention and requests are captured in some transcripts, but not in conversational turns. If you want the history of what you typed into the cli/tui you would have to datamine it from transcripts. By contrast, on the screenshots you can see it is a simple conversation that's saved and easy to return to.

- Multiplatform. As per the announcement "Use Codex across ChatGPT, your editor, and the terminal", meaning that you could open the ChatGPT app on your phone and continue to interact with this. While this might not matter to you if you exclusively code while sitting in front of your computer, still, it helps you be able to stay on top of things if you are on mobile.

- Background tasks and autonomous work. It says you can schedule it to do things in the background (issue triage, alert monitoring). At the moment, it would only do this if you've asked either codex or another agent to build a harness for it that does so. (i.e. you could ask Claude to build a harness that calls Codex on some schedule or interval.)

These all seem pretty good and I look forward to starting to use it once it's been more battle-tested.

[1] notably, on the Anthropic side the Opus 5 regressions - I had to downgrade to Opus 4.8 as I found Opus 5 was unreliable. I currently use Opus 4.8 and Codex Sol 5.6.

BodyCulture 9 hours ago||
This will make it much easier to trap people into installing it without any isolation just to their primary user account or even make it run with admin privileges by default.

With the recent security related events we already see that it seems to be wide spread practice that many operators avoid the hassle of isolating their agents.

It is just an incredible cultural experience to see how quick stupidity gets accepted as a behavioural default.

fg137 7 hours ago||
At least on Windows/Mac, a significant number of users of this app are non developers, and you should tame your expectations about them handling things like "isolation".

Even for developers working on Linux, you'll be surprised how many have never run a docker command in their life.

pjmlp 4 hours ago||
A proof that normies just can't handle security is that when OS X came to be, many Mac OS users used to ask how to disable the root/regular user dichotomy.

Or the long process to educate Windows users, and developers as well, to not run as Administrator.

usef- 9 hours ago||
The Linux version seems to use bubblewrap + seccomp for a "workspace sandbox" automatically, in addition to the auto mode and/or user approval.

A VM would be even better, but people could presumably run this in one if we manage to convince them of higher sandboxing in general ?

bloak 8 hours ago|||
Bubblewrap gets a bit confused inside a Docker container. If anyone knows how to run Codex inside a Docker container on Linux without it constantly complaining about a broken bubblewrap, please tell.
sh34r 1 hour ago|||
You have to add Linux capabilities to support creating nested namespaces. IIRC SYS_ADMIN is required. If you do that, what do you think you’re achieving with the container? It defeats the entire purpose IMO, just makes development more inconvenient.

If you need the sandboxing capabilities provided by the application and don’t want to run it on the host, you should set up a VM, not a container.

From a systems perspective, it’s foolish to rely on a proprietary application to provide its own isolation guarantees. This is an OS-layer responsibility.

ML/AI experts suffer badly from hubris. They are brilliant in one domain. It transfers poorly to systems engineering. The proof is in the FelonyBench results.

embedding-shape 8 hours ago|||
> If anyone knows how to run Codex inside a Docker container on Linux without it constantly complaining about a broken bubblewrap, please tell.

Why would you even use bubblewrap once it's running in a container? I've been running Codex in a container since it was publicly available, not even sure what you'd use bubblewrap for when you already separated codex from the rest of your filesystem?

ab71e5 6 hours ago||
Doesn't it automatically use bwrap if it's installed (in the container in this case)?
krzyk 4 hours ago||
Don't install bubblewrap? I didn't and use codex in docker.
bloak 2 hours ago||
"Codex could not find bubblewrap on PATH. Install bubblewrap with your OS package manager. See the sandbox prerequisites: ... Codex will use the bundled bubblewrap in the meantime."

I think neither the bundled bubblewrap nor the Ubuntu 24.04 bubblewrap works properly under Docker with the default options ... Yes, I've just tested that by asking Codex to "Test that bubblewrap is working properly" and then "Test again" after apt-get install bubblewrap.

EDIT: So the solution is to run "codex --dangerously-bypass-approvals-and-sandbox" in a Docker container?

justincormack 9 hours ago|||
The cli does that already with codex on linux
lucideer 7 hours ago||
The most remarkable things about this announcement:

- Electron based app: Electron is a framework sold on the basis of enabling rapid cross-platform development at the cost of performance.

- Frontier AI company: AI is sold on the basis of enabling rapid development

- App was released in February & took 6 entire months to port to Linux

SyneRyder 7 hours ago||
I guess it's easier to solve Erdos problems and improve the lower bound of the Riemann hypothesis, than it is to solve Linux desktop app distribution.
dingaling911 6 hours ago|||
LOL!

Though as someone who shipped commercial software on Linux, it wasn't that bad, you just had to ship everything other than the kernel.

boutell 5 hours ago||
LOL. Conveniently, electron is everything other than the kernel. Also, as much wasted RAM as an entire second instance of chrome that shares nothing, but oh well.
palmotea 4 hours ago|||
> LOL. Conveniently, electron is everything other than the kernel. Also, as much wasted RAM as an entire second instance of chrome that shares nothing, but oh well.

Well, with RAM being so cheap, why bother trying to trying to be memory efficient? It's not like OpenAI has an infinite pool of "AI" software engineers to work on this...

Oh wait...

dingaling911 4 hours ago|||
Yeah, but if you use Qt like I did, you're shipping libstdc++, some other crap and Qt.
timcobb 4 hours ago||
Yeah, I was kind of surprised recently that there's no solid native way to build desktop apps on Linux! I was thinking I'd build something actually native with AI but then the options were QT or or gtk...
preg_match 3 hours ago|||
Qt is as solid as it gets. It's very, very high quality. Much higher quality than Microsoft's frameworks on Windows.
timcobb 1 hour ago|||
Well yes, I was surprised by the situation in Windows too! So my stack is now split like this

- Apple: SwiftUI/AppKit

- Android: Jetpack

- Windows + Linux: Avalonia (C#)

Avalonia seems good so far. I couldn't do Qt because I can't deal with that qBittorrent look. GTK is aesthetically pleasing for me, but it only runs on Linux. Avalonia lets me target both of these UI-toolkit-less platforms with something native-ish. Flutter is probably good too, but I am using uniffi-rs and that doesn't generate Dart bindings AFAIK.

Would love to hear other options.

andai 1 hour ago||
Isn't Qt themeable? I also recall at one point they "emulated" native widget look and feel in their own renderer but I might be misremembering.
rubymamis 1 hour ago||||
Indeed! I wrote my note-taking app with a block editor (like Notion) in Qt with QML for the UI and C++ for the logic and it's been an incredible experience![1]

[1] https://rubymamistvalove.com/block-editor

timcobb 48 minutes ago||
Very nice, thank you for sharing!
cptskippy 3 hours ago|||
How would you know? It's impossible to have done a thorough analysis of all the Windows UI frameworks, five have been released since I started typing...
disiplus 4 hours ago||||
You don't care what were you trying to build but for a quick me alone Linux app I used Flutter. Honestly it's my go-to when I want to have a quick native app on any platform but dont want to bundle electron.
thewebguyd 2 hours ago||
Flutter is solid choice, and Canonical seems to be putting weight behind it for Ubuntu as well. Canonical is also now the new lead maintainer of Flutter Desktop.
blep-arsh 3 hours ago||||
QT and GTK are as native (and solid, I guess) as it gets, since there's no "default" UI toolkit that's shipped along with the kernel, unlike e.g. Windows or macOS. Speaking of Windows, why do they even ship browser-UI apps as a part of the OS? Do they find native Windows UI development too hard or feature-limited even with LLM assistance?
thewebguyd 2 hours ago||
Because Microsoft screwed up their native UI story. They are slowly starting to go native again with WinUI3, but Microsoft went through a rough period with UI frameworks, I mean how many do they have now all still "supported"? WinForms, WPF, MFC, WinUI, MAUI, etc.?
ilikehurdles 4 hours ago||||
Well what does native mean? An XWindow? Native on gnome is different from native on kde or xfce or lxde…
bpfrh 3 hours ago|||
I mean there is the no true linux argument, but covering the main stream desktops you basically have kde which is written with qt(so qt is native) or gnome which is written with gtk and both frameworks/DEs do implement other things besides "just" the gui, so you are native.
miohtama 4 hours ago||||
With the AI, we can finally have the year of desktop Linux.
OroPla 4 hours ago||||
Microsoft used to ship a Linux binary for Teams, but they stopped doing that. Guess the tech was lost.
freedomben 2 hours ago|||
I'm not usually one to defend Microsoft, but since Teams works great in a browser, I actually prefer it that way. My video conference doesn't need access to my whole filesystem and an opacity layer that prevents me from seeing what it does. I actually applaud them for supporting browsers as well as they do.
glitchc 3 hours ago|||
Yes, except it only worked some of the time on some of the distros (mainly Ubuntu/Debian variants of a certain vintage).
3abiton 3 hours ago||||
Nonetheless, it's the year og the linux desktop!
MichaelZuo 4 hours ago||||
Yeah idk what OpenAI was thinking. Just going off their own claims… OpenAI must be filled with bozos to not be able to utilize their own supposed product capabilities…
scrollop 4 hours ago||
Same with Anthropic and their "support" team, which seems to be Claude 0.1 run on a commodore 64 in the basement...
PunchyHamster 6 hours ago||||
It's the exact opposite. They picked the easiest way, done countless times just fine by everyone else and still failed
odiroot 6 hours ago|||
AppImage, yawn.
sedawkgrep 6 hours ago|||
I think they were joking...
cassepipe 5 hours ago|||
Yet I have to install a not very famous, maybe unmaintained daemon to get basic system integration
darkwater 7 hours ago|||
Exactly my thoughts! I was expecting a Rust app written from scratch available at least in Qt and GTK flavors , and maybe with a third and a fourth option.

Don't they have agent swarms with infinite tokens?

concinds 7 hours ago|||
They replaced a pretty decent native Swift Mac app (with only some chronic state management bugs), weighing ~100MB ish, with the Codex app that most people don't need, made in Electron, which takes minimum 5GB+ including filesystem cruft. Today I ran into a memory leak that froze the app and made the Renderer process take 5GB, then 10GB of RAM.

That was after I asked it to only touch 3 text files in its working directory, after which it sparked macOS permission prompts to access my Desktop, Documents, Downloads, and other apps' data. Thank god for the web platform and web apps.

lukan 5 hours ago|||
"Thank god for the web platform and web apps."

They can be amazing. But you can write shitty code in them like anywhere else.

jerf 3 hours ago|||
Browsers fundamentally chew on RAM for breakfast.

AIUI the reason is, "we", collectively even though we don't all work on them, optimize browsers for speed above almost all else. Then, when all the advertising websites on the internet consume all that speed, we can't go back and have to optimized for speed even harder. All that optimizing for speed ends up consuming lots of RAM, from all the images, the speed-optimize DOM with lots of cross-linking, the JIT and its generation of native code paths, the endless layers of caches, etc. etc. at every point where a browser can choose between RAM and speed it will choose speed unless it's just impossible to choose it for some reason.

So even if you write good code in Electron, it still tends to eat resources like candy.

On the plus side, if you write good code, you will pick up all those optimizations too, which is not necessarily all bad. Running something like VSCode at all is a powerful testament to the browser ecosystem now. IDEs do a lot of very heavy duty lifting and need a lot of performance to do all their work. But you can imagine that when the same text rendering system that enables all that fancy & fast syntax highlighting and all the other decorations and features in the editor window is used to display a preview of a markdown file that said preview is probably going to eat way more RAM than your instincts may have guessed.

concinds 5 hours ago|||
But they're strictly sandboxed and won't muck up my local filesystem.
kaycey2022 3 hours ago||||
What! They replaced the native mac app with this?!
rvz 6 hours ago||||
> They replaced a pretty decent native Swift Mac app (with only some chronic state management bugs), weighing ~100MB ish, with the Codex app that most people don't need, made in Electron.

Not even a $1T dollar company wants to maintain a native Swift Mac app even though they have AI models to build native apps for every platform.

Maybe they thought Electron would save them tokens to build the feature once and it works everywhere rather than spending more tokens on building the same feature on 3 platforms.

criddell 5 hours ago|||
Works everywhere except Linux for a long time.

What better way to showcase their tools capabilities than to deliver apps that are native, efficient, and integrate well with the host system? By compromising with Electron and making mediocre, bloated software they are signaling that they just aren't there yet.

It would be like Ford buying Hondas for their company cars. It might save them money but it sends the wrong message.

thewebguyd 2 hours ago||
Yeah that's what gets me. If, according to them, "software development is a solved problem" then there's no reason they can't make native, efficient, well integrated apps for each platform. If the problem is solved, just have the AI do it! If it's truly as advanced and accelerating as they claim, then there's no need for electron anymore.
zwirbl 6 hours ago|||
Maybe they are drinking their own kool-aid, resulting in less then optimal choices and worse software for everyone and everything
jonizzle 7 hours ago||||
I hate the new app aswell but you can still get the old one called ChatGPT Classic. Just scroll down to the bottom of the download site.
awakeasleep 6 hours ago||
Its unmaintained and has glaring bugs (cant adjust thinking level) for a month now
kilroy123 6 hours ago||
Yup, this pisses me off so much. I now just go to the web app instead of using this broken app.
mgambati 6 hours ago|||
It’s apple fault that swift and swift ui is completely garbage.
olavgg 6 hours ago||||
I have already created a croos-platform solitaire game in Rust with Opus 4.5. Done in one evening. I absolutely do not understand why OpenAI could not write a proper application in 6 months. Do they struggle to use the product they sell?
shuvrojit 5 hours ago||||
I am thinking about this too. I mean they have access to the highest level models and still they just produced a garbage.
hombre_fatal 3 hours ago|||
They probably just don't see it worth the extra effort. And maybe they want to keep it in sync with the web UI.

It definitely is a fraction of the work these days to maintain three different desktop apps + two different mobile apps + a web app. But it's still a lot more work than an Electron app that is merely tweaked between platforms.

randusername 4 hours ago|||
I was under the impression electron is chosen less because of broad cross-platform concerns and more because chat apps, specifically, require a level of care in text rendering that nobody really wants to design from first principles when they can just piggy back on web technologies.

But I wouldn't know first-hand.

lucideer 4 hours ago||
Text rendering is:

1. generally done well by many different UI frameworks

2. generally done better by most popular native UI frameworks than what Electron provides

BUT - those UI frameworks are typically not (very) cross-platform. In that context, Electron provides one of the best unified text-rendering stories of cross-platform options. But definitely far from the best overall.

yokoprime 6 hours ago|||
Wonder how many will bother using the desktop version on linux. Im on macos and i just use the CLI most of the time
jrflo 2 hours ago||
The desktop app is actually really good (esp on macos). I used to be CLI only, but after trying it recently it's super convenient. Much easier to manage multiple projects simultaneously and it has a built in diff viewer, file browser/reader, terminal, side chat interface, etc.
xbmcuser 7 hours ago|||
well this just shows how good the AI is at programming
sh34r 2 hours ago|||
The most remarkable thing is that it was built at all. What self-respecting Linux desktop user wouldn’t be satisfied with the TUI?

It is a marketing expense. We are talking about it. Surely the token cost was worthwhile on that basis alone. Students will likely get some usage out of it, too. But paying a human to do this port would have been an insane thing to do two years ago. I’d be surprised if they have more than 1000 MAUs for this client by the end of the year.

headrick 5 hours ago|||
How are you doing testing for desktop apps? any tool recommendation?
Muromec 7 hours ago|||
Corporate it is a scam pretending to be a job creation program to make sure people dont get funny ideas
bibimsz 4 hours ago|||
i doubt the UI was the hard part. these desktop apps are about computer use. thats where the tricky bits are.
sonu27 6 hours ago||
6 months is way too long IMHO
VCFundedGenYer 4 hours ago|
Nobody should be installing or using AI apps. Your browser is as far as they should go.

Reminder that the macOS app went from "this is fairly well made" to "it downloads multiple gigabytes of random things and doesn't follow any design conventions nor standards"

The Linux app will fail the same tests.

More comments...