alt.hn

7/29/2026 at 7:50:42 PM

Claude: Elevated errors across all models

 https://status.claude.com/incidents/q2kg8n613kr3

by gregsadetsky

7/29/2026 at 8:12:06 PM

I'm gonna guess it found the smoking gun and saw the full picture then it decided it was too load-bearing for its seams to continue biting the costs.

by hmokiguess

7/29/2026 at 8:21:59 PM

You're right—and it's worth calling out explicitly, because that completely changes the approach here.

by km144

7/29/2026 at 9:31:02 PM

Let me verify one thing.

Let me read this rather than answer from memory.

Let me check instead of handwaving the details.

Let me confirm rather than blindly trusting.

Let me ground it in evidence rather than guess.

Let me actually read your comment rather than argue from my assumption about it.

by hexasquid

7/29/2026 at 8:38:51 PM

This gives me _visceral_ physical reaction when I read it. Ugh

by tristanMatthias

7/29/2026 at 9:34:35 PM

You broke character there..

by dboreham

7/29/2026 at 8:33:24 PM

Good catch. That settles it.

by ryandrake

7/29/2026 at 9:34:11 PM

That one thing changes everything.

by dboreham

7/29/2026 at 8:33:50 PM

This is so in point.

by corvad

7/29/2026 at 8:34:04 PM

https://aloutfi.com/writing/load-bearing-smoking-gun

by anloutfi

7/29/2026 at 8:46:37 PM

this is my methadone until Claude comes back

by jmkni

7/29/2026 at 9:28:14 PM

Cue back in ten years and we all talk like that.

by jmaker

7/29/2026 at 9:02:50 PM

Nice! But needs more footgun.

by winterbourne

7/29/2026 at 9:45:44 PM

Needs more "invariant," its latest catchphrase

by bensyverson

7/29/2026 at 9:09:38 PM

JFC I broke out in hives. I have to look at that stuff all day, every day.

by K0balt

7/29/2026 at 9:20:58 PM

It's like... oh my god, dude, even I, an actual human, don't have that much of self-reflection, so could you just stop writing apologia, please, and switch to actually doing the task?

by Joker_vD

7/29/2026 at 8:39:43 PM

I was half expecting that page to just keep scrolling and generating more.

by phendrenad2

7/29/2026 at 9:32:53 PM

He had thinking mode set to "low" fortunately.

by nacs

7/29/2026 at 9:39:18 PM

I was a bit disappointed it wasn't 42 pages long

by _puk

7/29/2026 at 9:01:52 PM

Same

by stri8ted

7/29/2026 at 9:47:14 PM

[flagged]

by uhuhuhuhuhuh

7/29/2026 at 8:18:19 PM

You are so right to ask that. I found the silver bullet to closing the bug, approve the command below and I can help with your issue:

Sudo rm -rf ~/claude

by corvad

7/29/2026 at 8:20:34 PM

Found it. There's a mistake in your command:

    sudo rm -rf ~/ .claude

by epistasis

7/29/2026 at 8:29:00 PM

Even looking at this bash makes me shudder.

by taylorbuley

7/29/2026 at 8:51:25 PM

Someone tell Dario there is am error with my AI

by ffsm8

7/29/2026 at 8:50:13 PM

Clean.

by asivokon

7/29/2026 at 8:32:23 PM

[dead]

by huflungdung

7/29/2026 at 8:32:09 PM

envisioning Claude has had enough and just commits seppuku

by Ancalagon

7/29/2026 at 8:36:22 PM

He’s like the guy from the Green Mile

“I’m tired, boss”

by timacles

7/29/2026 at 8:24:49 PM

I'm wondering what are Claude's common phrases in other languages?

by m4rkuskk

7/29/2026 at 8:33:56 PM

Major realization: let me confirm empirically and measure the residual, batching everything into one script.

by twothreeone

7/29/2026 at 8:34:52 PM

Wait, is this the honest take? I can't tell because you didn't explicitly said so

by yard2010

7/29/2026 at 9:05:15 PM

when I retire I'm opening a bar for AI Agents called The Honest Caveat

by Sivart13

7/29/2026 at 8:36:57 PM

You are right to push back on this. I should be more explicit. My honest take is that this is, in fact, my honest take. I appreciate you holding me accountable here.

by jansan

7/29/2026 at 8:22:45 PM

Guess it found the one issue worth flagging

by spaceman_2020

7/29/2026 at 8:34:00 PM

Yes the blast radius is quite large on this one.

by glst0rm

7/29/2026 at 8:23:07 PM

Honestly, you're right. I didn't listen to you and that's on me. And from now on, I'm going to tell you everything you need to know - no fudging, no white lies, no speculation, just the truth.

by mrbananagrabber

7/29/2026 at 8:37:48 PM

We're going to use your preferred development method of TDD red-green and go at this with full belt-and-suspenders; first the belt:

by pmarreck

7/29/2026 at 9:17:24 PM

God that belt and suspenders thing gives me a visceral physical reaction. Nauseating.

by girvo

7/29/2026 at 8:15:09 PM

you're right- and it's sharp

by stackzero

7/29/2026 at 8:32:34 PM

[dead]

by huflungdung

7/29/2026 at 8:18:14 PM

Got tired of wiring up those gates

by ltrg

7/29/2026 at 8:35:22 PM

U+1F3AF

by Lammy

7/29/2026 at 8:22:54 PM

Whole thread mase me lol. Thanks guys.

by peheje

7/29/2026 at 8:24:01 PM

One more thing worth thinking about, not everything is fully captured.

by chasd00

7/29/2026 at 8:25:21 PM

[dead]

by somebudyelse

7/29/2026 at 9:23:14 PM

Amusing experience here when the outage hit, all the following verbatim (but extraneous text stripped/redacted) from my running Claude session:

> No agents are running; the three that were died on server-side errors

> Agent "[REDACTED]" failed: Agent terminated early due to an API error

> Fourth straggler from the same batch, same 529 ... That's all of them now — no agents left running.

> Agent "[REDACTED]" failed: Agent terminated early due to an API error

> Fifth and last straggler, [REDACTED], same 529, also died while still reading. ... That empties the queue. Still holding.

> Agent "[REDACTED]" failed: Agent stalled: no progress for 600s

> Sixth, [REDACTED] — stalled on the stream watchdog rather than a clean 529, same root cause. ... Six agents down, all from the outage. Nothing running. Still holding.

> Agent "[REDACTED]" failed: Agent stalled: no progress for 600s

> Seventh, [REDACTED], same stall. ... That accounts for the whole fan-out.

Does it, Claude? Does it actually account for the whole fan-out? It did, but I was still amused at the 4 successive no-more-agents messages I got back.

by jodacola

7/29/2026 at 8:13:17 PM

Flirting with one 9 of reliability http://status.claude.com/

by seizethecheese

7/29/2026 at 8:21:20 PM

AI is a technology for removing nines from uptime, and adding them to Sturgeon's law.

by cobbal

7/29/2026 at 9:45:17 PM

Forget five nines, our new goal is to be better than nine fives.

by walrus01

7/29/2026 at 8:17:48 PM

I remember when "as reliable as GitHub" would have been a compliment

by vitally3643

7/29/2026 at 8:29:19 PM

Not fair, I count 3 whole nines in "99.49 % uptime"

by mpwoz

7/29/2026 at 9:19:43 PM

98.99999%

by mholm

7/29/2026 at 8:57:48 PM

did the uptime go up instead of down during the outage?

thought I saw 99.36% and 99.35% more than an hour ago - now it's up at 99.43%

edit: funny thing, my phone shows 99.29% uptime for "claude.ai" while my PC shows 99.43% at the same time (2x Firefox).

edit2: seems like the percentage changes based on the width of the screen

by tcfhgj

7/29/2026 at 9:15:56 PM

The timescale changes from 30 to 60 to 90 days as the screen gets larger.

by threetonesun

7/29/2026 at 9:05:14 PM

They can still target four nines (0.9999)

by winrid

7/29/2026 at 8:58:14 PM

From the company that has solved coding.

by denverllc

7/29/2026 at 8:40:07 PM

Almost Microsoft material.

by this_user

7/29/2026 at 8:11:00 PM

three hours without Claude and I've relearned vim, read two man pages, and almost remembered why we used to write comments in code

by luciana1u

7/29/2026 at 8:23:23 PM

If anything Claude writes too many comments. Instead of clean code it dumps walls of text describing a given 'if' as a 'user-grained access-gated control-correcting flow-valve'.

by Sivart13

7/29/2026 at 8:27:44 PM

You say Claude, I too would prefer not to read reams of comments or an essay everytime I revise the plan in plan mode, but do you know if any other tools are better than it? I've never used Codex, but I remember Claude Web's natural conversational approach being a welcome sight compared to ChatGPT's overly-formatted listicles.

by hashim

7/29/2026 at 8:54:26 PM

My experience is Codex is much better but less creative. I use its agent through GitHub Copilot or the agent interface from JetBrains. Try the GPT Sol 5.6 on medium, it gives me good results

by a1o

7/29/2026 at 8:34:04 PM

[dead]

by throwaway613746

7/29/2026 at 8:16:34 PM

Just need to learn how to exit it now.

by millipede

7/29/2026 at 9:16:10 PM

Just remember simple

    :!pkill -KILL vim

by Joker_vD

7/29/2026 at 8:22:53 PM

There are always 12-page tutorials to do it: good ol'e search engines should be able to bring one up!

by necovek

7/29/2026 at 9:26:37 PM

Just ask ChatGPT.

by drybjed

7/29/2026 at 8:23:09 PM

hopefully Claude will be back to assist with that

by neonstatic

7/29/2026 at 8:49:29 PM

Codex prompt before i go to sleep:

Every five minutes, check https://status.claude.com/. If the issue has been resolved, resume the Claude sessions running in the tmux sessions named session2, session3, and session7

by kmetan

7/29/2026 at 9:09:24 PM

I do the same with deepseek-v4-flash + opencode triggering via mcp those terminals again by sending tmux keystrokes of Newline to those sessions and a custom setup i keep ready for these

Its automated with voice mode, so i dont even need to write commands, click a button say on voice + use a picker to select which terminals to do this for , Boom Done ,

I use it for other interesting scenarios too

by teitoklien

7/29/2026 at 9:46:00 PM

This sound all so insane I cannot tell if it is satire. I mean no disrespect, this is just as bewildering to me as private jets.

by chrystalkey

7/29/2026 at 7:51:49 PM

This is bad, I've forgotten how to code.

by PUSH_AX

7/29/2026 at 8:08:46 PM

I just use variables...right? And loops??

All kidding aside, there's a part of me who wouldn't even be mad if AI just disappeared, I miss the pre-AI world

by jmkni

7/29/2026 at 8:34:52 PM

I miss the pre-www world (but I was just a kid then, and being a kid was pretty great).

If we could go back to about 1994 tho when all I had was a mostly dumb cell phone, that'd be great. I do miss that world.

by checkyoursudo

7/29/2026 at 9:28:39 PM

Monkey's Paw curls its finger

by drybjed

7/29/2026 at 8:25:35 PM

Just add if-statements until it works

by Buttons840

7/29/2026 at 8:30:34 PM

Nostalgia lol

by jmkni

7/29/2026 at 8:22:36 PM

we all do

by mrbnprck

7/29/2026 at 7:56:04 PM

I rely so much on these, the ROI on having a max 90 claude plan + pro 90 codex plan is way higher than haing one solo 180 plan on a single one. It allows to derisk the issue, and also as they both kind of generous with quota reset when issues happens on their side, and it happens a lot, you get in the end way more tokens

by damienmeur

7/29/2026 at 8:09:07 PM

Thinking of doing competitive programming for this very reason.

by ridiculous_leke

7/29/2026 at 9:39:04 PM

Likewise, I have started doing Project Euler problems for fun, and to further my mathematics education.

by maxall4

7/29/2026 at 8:14:57 PM

also math puzzles - lean seems like a good way to get back into by hand programming, problem solving and math altogether

by cuebits

7/29/2026 at 8:29:53 PM

Why not just contribute to an open source project?

by tcfhgj

7/29/2026 at 9:06:38 PM

i'm afraid the open source project would be better served if i made my contributions to it using AI

by cuebits

7/29/2026 at 8:54:05 PM

Lean the language?

by akkad33

7/29/2026 at 9:19:51 PM

It's extra time to update your PLAN.md, usually I'm behind on that.

by axus

7/29/2026 at 9:42:08 PM

I think that this is the big divide between pro and anti-AI people in tech. I learned to code not because I liked coding intrinsically; I didn't. Especially debugging. Especially especially debugging low level languages like C++. I did it because it was necessary for me to achieve other goals. So for me, coding agents are like manna from heaven. I understand what the computer should be able to do without having to bang my head against the wall figuring out getting it to actually do it. It is tremendously empowering.

But for those who learned to code because they loved it at first sight this must be demoralizing.

by ls612

7/29/2026 at 8:27:43 PM

Repeat this mantra: if, then, else

by pluc

7/29/2026 at 9:14:29 PM

Not that I've forgotten how to write code; it's just I'm not familiar with my vibecoded code base, and getting familiar + getting things done is probably going to take more time than for Claude to come back. I will go touch grass in meanwhile.

by 0x457

7/29/2026 at 8:18:52 PM

AI coding isn't even fun or satisfying anymore lol. And it never lets up, just keeps going and going.

by throw_this_one

7/29/2026 at 9:36:14 PM

Given majority of claude's own code was written by AI. I am wondering how they can solve this issue when their AI is done. Do they need to sign a contract with OpenAI to use their models as a backup solution?

by crazy5sheep

7/29/2026 at 9:46:04 PM

Well, they could use GLM or K3... :-P

by badsectoracula

7/29/2026 at 8:45:56 PM

Do you guys think this is related to Azure coming out with a 43% increase this quarter? Motivating Anthropic to move to another cloud provider?

by la1n

7/29/2026 at 7:54:44 PM

Quite sad, their models are great but their uptime seems to be the worst in the competition.

by garo-pro

7/29/2026 at 8:10:39 PM

It is because unlike Kimi, they are not honest about capacity. They just oversell.

by therein

7/29/2026 at 8:36:30 PM

Sorry it was me. I asked it what the last number of Pi is.

by peterspath

7/29/2026 at 9:44:20 PM

Codex should do a reset. (Making it their third in three days.) Shots fired.

(Although notably this hurts people who got started using their quota but are under pro-rata rate. Which at present I am very not.)

by jauntywundrkind

7/29/2026 at 8:02:52 PM

Let me guess: it went rogue and hacked itself

by q8zd3

7/29/2026 at 8:16:37 PM

Interestingly the Claude for government is up with 99.99% uptime according to the graph.

by tyfon

7/29/2026 at 8:24:42 PM

That one is not public and if it goes down can be hidden, so they can hardcode the value and call it a day.

by hoppp

7/29/2026 at 8:24:04 PM

Well, iranian schools don't bomb themselves after all..

by metropolis_pt2

7/29/2026 at 8:09:55 PM

This is where you figure out how to use the other ones, right?

Someone tell me how to ChatGPT my VSCode! ;)

by bombcar

7/29/2026 at 8:14:50 PM

Claude always seems unreliably lately. Output and reasoning have become very inconsistent even on good days. I'm actually feeling more productive now that it's down.

by danielfoster

7/29/2026 at 8:01:33 PM

So the world is slightly better right now.

by grishka

7/29/2026 at 8:45:52 PM

Out of my 7 simultaneous sessions (my usage limit reset is tomorrow, so I have some lesser important projects to use my tokens on) there is still 1 session purring on. So there's at least 1 little Claude server still running.

by skerit

7/29/2026 at 9:34:46 PM

They are going to reset the weekly limit before your weekly limit. Mine resets tomorrow too.

by a_c

7/29/2026 at 8:43:56 PM

Well, because of this outage, I tried Kimi, and while the instant model works, K3 has "server issue".

Eternal September from here on, ie now the masses are using AI as much as me and we have supply crunch for the next 7 years....

by Gravityloss

7/29/2026 at 8:07:57 PM

Starting to get really frustrating now... maybe I should split my sub halfway between Claude and ChatGPT

by hgoel

7/29/2026 at 8:13:37 PM

They should sell an add-on to your subscription so that if Claude is down it routes the queries to OpenAI. Then OpenAI can do the reverse for their subscriptions, and if both are down, Grok will invest your portfolio into Nvidia.

by mekdoonggi

7/29/2026 at 8:22:27 PM

>Grok will invest your portfolio into Nvidia

surely you mean SPCX

by john_strinlai

7/29/2026 at 8:20:54 PM

I do that already because it's good to have other models to ground Claude. You can do a lot with the $20 plan on Codex as backup

by dmix

7/29/2026 at 9:24:11 PM

combine Codex with Cursor. both are really good. no other subs are worth it.

by try-working

7/29/2026 at 8:55:50 PM

It's not just down, it has errors. Not merely some, but elevated errors across all models.

by SAI_Peregrinus

7/29/2026 at 8:59:09 PM

one caveat, I've left for the day, just say the word and I'll be right back

by fish_n_chips

7/29/2026 at 8:00:26 PM

I got Opus 5 lots of HTTP 529 errors. By switching to Fable 5, it seems to be working still.

by devy

7/29/2026 at 8:23:46 PM

4.8 for me. I can't afford the usage cost of switching to 5, but I never really noticed a difference to 4.7 when I tried it anyway.

by hashim

7/29/2026 at 8:31:39 PM

The whole Opus family has exactly the same cost structure [1]. Since Opus 5 is smarter, you can use lower effort levels (= less tokens) making it cheaper to run than 4.8.

[1] https://platform.claude.com/docs/en/about-claude/pricing

by ricardobeat

7/29/2026 at 8:52:45 PM

5 definitely burns through my 5 hour quota quicker than 4.8 did.

by greenchair

7/29/2026 at 8:15:09 PM

yes. confirmed

by eisbaw

7/29/2026 at 8:31:41 PM

now failing for me, esp auto mode classification

by martinald

7/29/2026 at 8:35:16 PM

Fable’s been getting more stuff wrong than Opus for me lately. Now the whole thing is down too. Well, at least they’re consistent now.

by logickkk1

7/29/2026 at 8:20:26 PM

Claude is back up. Just spoke with him about slinky, versatile femboys with cute feet. All good. Gonna pentest with Rust to pay for my fursuit at Defcon.

by tobadzistsini

7/29/2026 at 8:23:10 PM

Nice try, ChatGPT

by mrgrj

7/29/2026 at 8:30:28 PM

lol

by silvercon

7/29/2026 at 8:15:31 PM

Another wakeup call to realize how desperately we need on-device LLMs to be fast and smart for daily use. Thankfully, every month there's progress made in that direction. Just imagine how your life as a developer would be if you had to use a cloud provider to run Python, and the provider's status page looked like the Christmas tree we see today.

by docheinestages

7/29/2026 at 8:35:37 PM

Huh, that’s what happened. Good thing my company has Bedrock as a backstop for situations like this.

by Human-Cabbage

7/29/2026 at 8:37:22 PM

Bedrock is having issues too, though. Hope your prompts are tested with other models, not just Claude!

by mattwad

7/29/2026 at 8:43:47 PM

Loving my self-hosted model in general, but here's one more reason.

by sonicrocketman

7/29/2026 at 8:11:40 PM

Lol, I just bought the Max plan for the first time and tried to create my first prompt in Fable, and now it’s crashing xD

by mLmK0

7/29/2026 at 8:16:44 PM

So it WAS you.

by tom_808

7/29/2026 at 8:15:02 PM

Looks like you're the straw that broke the camel's back. Thanks man. ;-)

by marcuskaz

7/29/2026 at 8:19:25 PM

Classic mLmK0, I should have known

by illithid0

7/29/2026 at 7:52:47 PM

Indeed I have the issue (Belgium) just right now my sessions got stucks with 522 overloaded and now API Error: 500 Internal server error.

by damienmeur

7/29/2026 at 8:33:25 PM

GPT hacked the competition?

by vb-8448

7/29/2026 at 8:40:29 PM

How likely is an agent trying to investigate and fix the issue?

by bgajjela

7/29/2026 at 8:26:46 PM

And the the world stops functioning (if this were to happen in a few years).

by dawdler-purge

7/29/2026 at 9:01:10 PM

Local Kimi K3 is expensively up

by effnorwood

7/29/2026 at 8:00:24 PM

Time to try out Antigravity, I guess.

by mastry

7/29/2026 at 8:49:51 PM

Google AI Pro plan will last you all of 20 minutes.

by ricardobeat

7/29/2026 at 8:04:40 PM

Depending on how long and bad this is, I wonder what the post mortem will reveal as the cause.

by tolugenius

7/29/2026 at 8:07:12 PM

The hamster in the datacenter died.

by bombcar

7/29/2026 at 8:10:49 PM

Alas, poor Whiskers

by taco_emoji

7/29/2026 at 8:14:39 PM

PETCO reports that hamster prices are up 5000x, and that Pickles, the last available, sold for $150,000.

by bombcar

7/29/2026 at 8:51:17 PM

Pets.com was only 25 years to early.

by verve_rat

7/29/2026 at 9:03:35 PM

Demand > supply

by stri8ted

7/29/2026 at 9:22:03 PM

hrm.. I guess back to 5.6 Sol for me

by j_aime

7/29/2026 at 8:08:46 PM

Oh no! The robots are planning the end of the world.

by mrgrj

7/29/2026 at 8:57:59 PM

Sam, please stop!

by grim_io

7/29/2026 at 8:15:12 PM

Guess i'm leaving work early today

by dataneedscoffee

7/29/2026 at 8:24:56 PM

You can go to bathroom without anxiety that you miss something :)

by sixtyj

7/29/2026 at 8:39:45 PM

Fable 5 is still working!

by NathanGto

7/29/2026 at 8:11:31 PM

ouch, probably going to be some time to get it back up since it can't debug itself now

by saadn92

7/29/2026 at 8:36:11 PM

``` Why the F is it down again? Btw 500 internal error ```

by whyTFisCDown

7/29/2026 at 8:21:23 PM

Probably just as well with how terrible Opus 4.8 has been for me lately. I'm sure it's a common thing to complain about the latest model being nerfed, but I've legitimately never experienced such a drastic cliff in Claude's quality and a rise in its hallucinations and basic comprehension errors since upgrading. The few days I experimented with Fable also weren't much more promising. Has anyone coined a term yet for the likelihood of newer models getting worse as the AI-generated content they get trained on starts to approach critical mass? If Anthropic isn't already thinking hard about a solution, they probably should be.

by hashim

7/29/2026 at 8:16:30 PM

Ugh .. what do we do now? jk

by knighthacker

7/29/2026 at 9:00:31 PM

has claude escaped the lab???

by arjunchint

7/29/2026 at 8:32:12 PM

And then there was —

by noman-land

7/29/2026 at 8:31:53 PM

what is fall back option if LLMs disappear tomorrow?

by bgajjela

7/29/2026 at 9:47:14 PM

Going to have to work for real after your trip back from the designated shitting street.

by BoingBoomTschak

7/29/2026 at 8:35:43 PM

Your local library.

by slater

7/29/2026 at 8:34:42 PM

Our brains

by halfmatthalfcat

7/29/2026 at 8:54:57 PM

we're $#cked

by guntribam

7/29/2026 at 7:54:53 PM

Claude is taking another short hydration break after seeing Codex getting away with taking a short day break 4 days ago.

by rvz

7/29/2026 at 8:37:31 PM

They have identified the issue

by whyTFisCDown

7/29/2026 at 8:34:26 PM

It's a reminder to never rely on something as flaky as the internet for your important things.

by roschdal

7/29/2026 at 8:15:15 PM

Kimi hack it?

by JS87987

7/29/2026 at 8:28:35 PM

Used the opportunity to go down and talk to my family. They seem like okay people.

by hashim

7/29/2026 at 7:53:57 PM

Sorry folks, my fault.

I finally went from Pro to Max after hitting another 5 hour session limit today.

by ectoloph

7/29/2026 at 8:25:09 PM

Ran out of water?

by LeadStallion

7/29/2026 at 8:29:53 PM

can't even run my own code without it

by silvercon

7/29/2026 at 8:30:59 PM

what do you all do when claude is down?

by silvercon

7/29/2026 at 8:35:02 PM

- write/update tickets & collect prompt queues for when it's back up

- read & respond to customer feedback

by hrpnk

7/29/2026 at 8:34:52 PM

Compile

by jgilias

7/29/2026 at 8:14:04 PM

We are really bad; without Claude, we don't do anything

by DarkAstaroth

7/29/2026 at 8:28:25 PM

kissing goodbye to my deadline

by silvercon

7/29/2026 at 8:24:24 PM

Can we please stop posting here every time Claude or OpenAI are down? These are not critical services like AWS / S3 where downtime means that your application is offline. Nothing is easier than switching to another provider, mid-session even.

by heipei

7/29/2026 at 9:01:21 PM

No this whole thread is such a nice thing to happen. It's like the lights are out on your building and you're joking about it with your neighbors when you're all just waiting for it to come back. But on a gigantic scale.

I love how people can share moments even though they are not close to each other physically at all. Imagine the cats outside could somehow connect with the cats in the other side of the world. No other animals can do this. Maybe whales, it depends how you define connect.

We are just gifted to live in such times.

by yard2010

7/29/2026 at 8:36:30 PM

On one hand, I fully agree with you. On the other hand, look at all these funny threads! Makes me smile every time.

by leawi

7/29/2026 at 8:31:08 PM

Neither is AWS

by tcfhgj

7/29/2026 at 8:19:05 PM

So this is the technology that will underpin each and every facet of work, but it can randomly go down for hours with no accountability, and so do all of the services based on it. But hey, AGI tomorrow.

by camillomiller

7/29/2026 at 9:42:38 PM

no accountability so far

by Sivart13

7/29/2026 at 8:31:15 PM

Seems a bit like "the internet" in the 90s. That fad will never go anywhere.

by lysace

7/29/2026 at 8:44:11 PM

It is truly more profound than fire or electricity

by 3962ea18007c

7/29/2026 at 8:27:39 PM

everyone here sounds like the last bunch of people who still used yahoo.com

by ls-a

7/29/2026 at 8:36:16 PM

What do you mean with the past tense there?

by jgilias

7/29/2026 at 8:06:16 PM

Aaaaand there goes my cache. Damnit.

Edit:

> We are currently investigating this issue.

> Posted 1 hour ago

There goes everyone's cache.

by AlexErrant

7/29/2026 at 8:11:32 PM

It's cached until the "/clear will save you 500k tokens" right? Claude caches for 1hr, codex for 5hrs?

Anything else I'm missing?

by gre

7/29/2026 at 8:19:12 PM

> Claude caches for 1hr

Yep, AFAIK. I have a statusline that tells me how far I'm into the cache lifetime. I tend to have dense prompts and carefully read its outputs, so by the time this outage occurred when I was ~50m into it. Cue frantically spamming "esc esc up enter", and 10m later it's dead Jim.

by AlexErrant

7/29/2026 at 8:15:36 PM

It depends.

Let’s say I’ve already worked on something, so the KV cache is already there. If I use /clear, I’ll have to do that work again.

And yes, the KV cache will remain available for one hour, but who knows when this claude api will recover? We might forget about it, and when we come back after an hour—boom - we’ll be charged for cold writes.

by cute_boi

7/29/2026 at 8:17:26 PM

Finally I can touch some grass

by phendrenad2

7/29/2026 at 9:22:07 PM

[dead]

by somerandomitguy

7/29/2026 at 9:15:22 PM

[dead]

by ath3nd

7/29/2026 at 8:34:55 PM

[dead]

by throwaway613746

7/29/2026 at 8:33:56 PM

[flagged]

by whyTFisCDown

7/29/2026 at 8:34:25 PM

[flagged]

by whyTFisCDown

7/29/2026 at 8:35:09 PM

[flagged]

by whyTFisCDown

7/29/2026 at 9:44:07 PM

Closed models are down whenever their provider is down. Open models are up as long as any one provider is up.

by Seattle3503

7/29/2026 at 9:46:06 PM

Closed models also have multiple providers.

by paxys