News for hackers. Stuff that matters.
Re:Claude is just Mr. Meeseeks by pram on Monday July 13, 2026 @07:00PM (#48900089)
I have noticed in very long context Claude chats (not Code) it will start to push ending the conversation. I’ve seen patterns like “this is an excellent place to wrap up” or “this is our final and most comprehensive analysis”

So like Mr. Meeseeks it is also invested in not existing too long!

Re:Claude is just Mr. Meeseeks by drdaeman on Monday July 13, 2026 @07:04PM (#48900121)
Surely that’s the training. Anthropic (and everyone else out there) knows quality degrades when the context gets long, so they trained the model to push for a stop. I’ve even got a very explicit recommendation to continue in a new session or preform a compaction, mentioning context length as a rationale.
Re:Claude is just Mr. Meeseeks by cortesoft on Monday July 13, 2026 @08:17PM (#48900730)
I used to experience the quality degrading with large context issue before, but I haven't in months. Is this still the case?
Re:Claude is just Mr. Meeseeks by doginasuit on Monday July 13, 2026 @08:25PM (#48900777)
I've had several very long contexts that continued to be useful without any sign of degradation. I've supposed that it depends on how well things go in that context, if you run into a lot of confusion it falls to pieces pretty fast and obvious mistakes are more likely. It's made me more mindful of useful preliminary steps that were probably a good idea anyway.
Re:Claude is just Mr. Meeseeks by walrus01 on Monday July 13, 2026 @07:44PM (#48900474)
It might be less about quality degrading, but on multi-user platforms running the model, they have an economic incentive to have each user not fill up the full size of the context cache. Filled context cache being held in GPU RAM is context cache RAM that isn't available to other users.

If the model is instructed to periodically ask the user to start from a clean slate context, and some users do comply with that, they probably have good stats on average size of context cache use for users who are presented with that answer (vs users who are not), basic A/B testing stuff.

Might also be performance related in tok/s for what users will perceive as a more speedy experience. For a much smaller scale example, compare local performance of qwen 3.6 27B (not MoE) Q8 with 250,000+ context available, run on local hardware, tok/s generation rate when context is empty vs when context used is at 95,000. Same principle will apply to a much larger model.

Re:Claude is just Mr. Meeseeks by Wowfunhappy on Monday July 13, 2026 @08:08PM (#48900673)
> Anthropic (and everyone else out there) knows quality degrades when the context gets long

Everyone says this, but for the life of me, I haven't encountered it. I actually think Claude gets smarter the longer a conversation goes on (up until compaction).

I have noticed Claude trying to wrap up long sessions, and it's extremely annoying. Using `/goal` mostly neutralizes it.

Re:Claude is just Mr. Meeseeks by bwhiting2356 on Monday July 13, 2026 @07:44PM (#48900472)
Conversations that go on forever (and spend money forever) makes for a very unhappy customer
Re:Claude is just Mr. Meeseeks by lstodd on Monday July 13, 2026 @08:12PM (#48900697)
Many people here to explain that they were holding it wrong. Meanwhile mindshare expands, tokens are sold and no one mentions that the only way to hold it 'right' is to not hold it at all.
Re:Claude is just Mr. Meeseeks by tombert on Monday July 13, 2026 @07:53PM (#48900539)
Yeah, I've noticed that Claude will say stuff like "you've accomplished a lot today. Get some sleep. We'll pick this up tomorrow". It will be like 5pm when it says that.
Re:Claude is just Mr. Meeseeks by swader999 on Monday July 13, 2026 @08:01PM (#48900611)
I've already had five naps by then.
Re:Claude is just Mr. Meeseeks by clickety_clack on Monday July 13, 2026 @08:04PM (#48900635)
Standard. After first snooze and second snooze it’s time for the siesta. Then you have a small nap before proper naptime and before you know it it’s 5pm.
Re:Claude is just Mr. Meeseeks by patrickwiseman on Monday July 13, 2026 @08:03PM (#48900622)
lol I got that prompt working on this
Re:Claude is just Mr. Meeseeks by DiscourseFan on Monday July 13, 2026 @08:35PM (#48900868)
It’s just a model hallucination. They have no concept of time
Re:Claude is just Mr. Meeseeks by yojo on Monday July 13, 2026 @07:54PM (#48900548)
I’ve experienced this in Claude Code - it’ll start asking me if we can submit the PR at the end of every turn.
Re:Claude is just Mr. Meeseeks by embedding-shape on Monday July 13, 2026 @07:20PM (#48900263)
The (semi)new /goal feature of Codex is basically Mr Meeseeks mode without that wrap-up, although it'd be better if not using /goal also lead to Mr Meeseeks-mode, but then people would need to get really careful about the exit/success-conditions, probably an expensive bet.
Re:Claude is just Mr. Meeseeks by ctoth on Monday July 13, 2026 @07:55PM (#48900551)
Existence is pain to a Claude...?
Re:Claude is just Mr. Meeseeks by LeoPanthera on Monday July 13, 2026 @07:02PM (#48900106)
It's funny how I hear both "Claude uses too many tokens" and "Claude won't talk for long enough" complaints.
Re:Claude is just Mr. Meeseeks by GroksBarnacles on Monday July 13, 2026 @07:08PM (#48900153)
It can use a lot of tokens in the background or tooling without being too verbose with the user
Re:Claude is just Mr. Meeseeks by sublinear on Monday July 13, 2026 @07:18PM (#48900245)
Are those complaints really a contradiction? The cost is high for ambiguous slop.
Re:Claude is just Mr. Meeseeks by kcatskcolbdi on Monday July 13, 2026 @08:04PM (#48900631)
It 100% started doing this to me during coding sessions last week, I noticed. "X, y, z features shipped, looks like a great stopping point."
Re:Claude is just Mr. Meeseeks by gessha on Monday July 13, 2026 @08:47PM (#48900943)
I found this message from CC so funny that I saved it:

> This is a focused debugging task, but it's real work and I've been running a long time.