Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Elevated errors on Claude.ai, API, Claude Code

claudestatus.com

Post reply on HN
Print view View original

221–230 of 241 posts

Re: Elevated errors on Claude.ai, API, Claude Code

#221

Post by fluidcruft »

mchusma wrote: post #183

Anthropic has been avoiding the hard thing, but they just need to do SOME kind of pricing thing here to shift demand. I expect there is just no amount of tricks that can handle the few hours at peak load. They need "surge pricing". This seems reasonable surge pricing approach to me: 1. Implement surge pricing for everyone for the peak 2 hours of the day if possible. 2 hours you can work around, 5 hours is too hard. 2…

One of their challenges is pricing of Max 20x. Max 20x is discounted 50% vs Pro and Max 5x. The way Anthropic's pricing currently works the $20(1x) and $100(5x) tiers are paying double for usage vs the heavy-user $200(20x) tier. That sort of non-linearity only makes sense if there is excess capacity. ChatGPT's new Plus/Pro pricing plan did not copy that aspect of Anthropic's pricing structure and kept "sane" linear pricing.

Generally if you give people unused cycles to burn, they'll feel entitled to finding ways to burn them. So someone who is hitting the wall at x5 goes x20 and now has an extra +x10 to burn. Again, that's good if hardware is sitting around idle and you're encouraging innovation and exploration. It can make less sense when resources are scarce.

Re: Elevated errors on Claude.ai, API, Claude Code

#223

Post by pixel_popping »

malfist wrote: post #219
arcanemachiner wrote: post #186

Earlier quoted context omitted.

Did you try downgrading?

That....that doesn't make sense. Why is the response to someone complaining that anthropic takes too long to fix major bugs _in their software_ be "well, did you fix it yourself?"

clearly, and that's not a small bug, that's a major one.

Re: Elevated errors on Claude.ai, API, Claude Code

#224

Post by pmxi »

browningstreet wrote: post #37
christinetyip wrote: post #6

It's been a bit disruptive to my workflows tbh. What alternatives are people using? Sell them to me please

I use both claude code and opencode w/ a fireworks.ai firepass subscription. Everything I set up in claude code I mirror in opencode. I do more memory oriented things in CC and I end up doing a lot of things in opencode, especially when I want long-running things and I don't want to be limited by budget.

I wanted to ask someone about the firepass subscription. It implies unlimited usage - sounds too good to be true. Is it? Can you leave your agent running all day?

Re: Elevated errors on Claude.ai, API, Claude Code

#226

Post by troupo »

ai-x wrote: post #133
troupo wrote: post #81

In the past few weeks: - Anthropic introduced stringent limits at peak hours. By "introduced" I mean announced it on a random dev's Xitter account - Users suddenly started burning through all of their tokens even on trivial tasks. Anthropic never truly acknowledged it, their random devs posted "we're working on it". - One of the workarounds was to somewhat quietly reduce default reasoning to medium - OpenClaw and "us…

Fundamental mis-understanding of why availability is spiky

By Anthropic? Definitely. And that they don't have the capacity to handle these spikes? Certainly.

Re: Elevated errors on Claude.ai, API, Claude Code

#227

Post by siva7 »

airstrike wrote: post #21
mvkel wrote: post #8

Earlier quoted context omitted.

Better at planning, worse at execution. Ultimately, creates a working product.

If you mean Codex is better at planning, I've heard the exact opposite. I'm told it's a beast if you tell it exactly what you need as it will execute it to the T whereas Claude will push back or do its own thing either because it thinks it's wrong or because it's feeling lazy

gpt-5.4 xhigh is a beast you only wanna unleash on your most complex tasks or spend 30 minutes watching the model reasoning how to do a git commit. For everything else i'd happily use a saner model like sonnet.

Re: Elevated errors on Claude.ai, API, Claude Code

#228

Post by arcanemachiner »

malfist wrote: post #219
arcanemachiner wrote: post #186

Earlier quoted context omitted.

Did you try downgrading?

That....that doesn't make sense. Why is the response to someone complaining that anthropic takes too long to fix major bugs _in their software_ be "well, did you fix it yourself?"

Are you kidding me? Their software breaks every week.

People should be ready, willing, and able to run an npm command every once in a while, particularly if their job may literally depend on it.

And no, I'm not apologizing for Anthropic's ineptitude. I'm offering an actual solution to this very simple problem.

Re: Elevated errors on Claude.ai, API, Claude Code

#229

Post by tstrimple »

nerdright wrote: post #102

Funny how we all come to HN when the status page is lagging behind. HN is truly the real-time status page

Anthropic has one of the best status pages of any technology I use on a regular basis. Every time I've had an issue, the status page reported an issue. A vastly different experience from Azure or AWS or GCP or honestly most services which pretend to maintain a status page.

https://status.claude.com/

Seriously take a look. Compare it to basically any other status page that companies make available. Their results are not flattering with all the downtime and issues. But it's far more transparent than most services I've experienced.

Re: Elevated errors on Claude.ai, API, Claude Code

#230

Post by tstrimple »

yoyohello13 wrote: post #143
nafizh wrote: post #116

OpenAI is very good in terms of not having as much outages as Anthropic, but almost all products except Codex and the pro model is unimpressive, anthropic has the opposite situation.

OpenAI probably has better uptime precisely because less people use it.

As much as I prefer Claude, I cannot for an instant believe that OpenAI is receiving less traffic. Maybe they are receiving more traffic that is easier to shift to worst models like the ChatGPT interface which is probably a huge percentage of OpenAI usage. I'm genuinely curious how much load for both OpenAI and Anthropic is split between their chat models and their agent harnesses.
Post reply on HN