News for hackers. Stuff that matters.
Re:Claude Code Daily Benchmarks for Degradation Tracking by samlinnfer on Thursday January 29, 2026 @06:56PM (#46818709)
Is there compensation for the tokens because Claude wasted all of them?
Re:Claude Code Daily Benchmarks for Degradation Tracking by jonplackett on Thursday January 29, 2026 @07:27PM (#46818987)
So quiet…
Re:Claude Code Daily Benchmarks for Degradation Tracking by TZubiri on Thursday January 29, 2026 @09:41PM (#46819950)
It is possible that degradation is an unconscious emergent phenomenon that arises from financial incentives, rather than a purposeful degradation to reduce costs.
Re:Claude Code Daily Benchmarks for Degradation Tracking by gizmodo59 on Thursday January 29, 2026 @09:50PM (#46820003)
Codex seems to give compensation tokens whenever this happens! Hope Claude gives too.
Re:Claude Code Daily Benchmarks for Degradation Tracking by mathrawka on Thursday January 29, 2026 @10:02PM (#46820068)
You are funny. Anthropic refuses to issue refunds, even when they break things.

I had an API token set via an env var on my shell, and claude code changed to read that env var. I had a $10 limit set on it, so found out it was using the API, instead of my subscription, when it stopped working.

I filed a ticket and they refused to refund me, even though it was a breaking change with claude code.

Re:Claude Code Daily Benchmarks for Degradation Tracking by TOMDM on Friday January 30, 2026 @01:13AM (#46821104)
Anthropic just reduced the price of the team plan and refunded us on the prior invoice.

YMMV

Re:Claude Code Daily Benchmarks for Degradation Tracking by MichaelZuo on Friday January 30, 2026 @02:23PM (#46828709)
So they have no durable principles for deciding who or what to refund… doesnt that make them look even worse…?
Re:Claude Code Daily Benchmarks for Degradation Tracking by necovek on Saturday January 31, 2026 @01:30AM (#46834037)
Or they do, and two sentences from two different experiences don't tell a full story?
Re:Claude Code Daily Benchmarks for Degradation Tracking by MichaelZuo on Saturday January 31, 2026 @10:28AM (#46837481)
Okay “they do” based on what more compelling evidence?

Its not like the credibility of the two prior HN users are literally zero…

Re:Claude Code Daily Benchmarks for Degradation Tracking by necovek on Saturday January 31, 2026 @11:59AM (#46838340)
I am saying there is no evidence either way: they had contrasting experiences and one GP established this means that company has no standardized policies. Maybe they do, maybe they don't — I don't think we can definitively conclude anything.
Re:Claude Code Daily Benchmarks for Degradation Tracking by MichaelZuo on Saturday January 31, 2026 @01:37PM (#46839321)
So if you acknowledge the prior claims have more than literally zero credibility… then what’s the issue?

That I dont equally weigh them with all possible yet-to-be claimed things?

Re:Claude Code Daily Benchmarks for Degradation Tracking by necovek on Tuesday February 03, 2026 @05:25AM (#46869141)
I object to your conclusion that "they have no durable principles": not sure how do you get to that from two different experiences documented with a single paragraph.
Re:Claude Code Daily Benchmarks for Degradation Tracking by MichaelZuo on Tuesday February 03, 2026 @06:00PM (#46878643)
Because I can assess things via probability… without needing 100% certain proof either way?
Re:Claude Code Daily Benchmarks for Degradation Tracking by necovek on Wednesday February 04, 2026 @02:36PM (#46890553)
This is becoming futile: this is not even about proof, but there not even being a full account of two cases you are basing your opinion on.

Obviously, you can derive any opinion you want out of that, but while I am used to terms like "probability" being misused like this, I've generally seen a higher standard at HN.

To each their own, though. Thank you for the discourse and have a good day.

Re:Claude Code Daily Benchmarks for Degradation Tracking by mvandermeulen on Friday January 30, 2026 @04:23AM (#46822262)
You’re lucky they have even admitted a problem instead of remaining silent and quietly fixing it. Do not expect ethical behaviour from this company.
Re:Claude Code Daily Benchmarks for Degradation Tracking by port11 on Friday January 30, 2026 @07:34AM (#46823735)
Why not, can you expand? Asking because I’m considering Claude due to the sandbox feature.
Re:Claude Code Daily Benchmarks for Degradation Tracking by caspar on Friday January 30, 2026 @11:11AM (#46826126)
FYI the sandbox feature is not fully baked and does not seem to be high priority.

For example, for the last 3 weeks using the sandbox on Linux will almost-always litter your repo root with a bunch of write-protected trash files[0] - there are 2 PRs open to fix it, but Anthropic employees have so far entirely ignored both the issue and the PRs.

Very frustrating, since models sometimes accidentally commit those files, so you have to add a bunch of junk to your gitignore. And with claude code being closed source and distributed as a bun standalone executable it's difficult to patch the bug yourself.

[0]: https://github.com/anthropic-experimental/sandbox-runtime/is...

Re:Claude Code Daily Benchmarks for Degradation Tracking by port11 on Monday February 02, 2026 @07:52AM (#46855445)
Hmm, very good point indeed. So far it’s behaved, but I also admit I wasn’t crazy about the outputs it gave me. We’ll see, Anthropic should probably think about their reputation if these issues are common enough.