𝖓𝖎𝖈𝖔𝖉𝖊𝖒𝖚𝖘 🏴

illúvatar akbar

@Kaylara0wl If you think a millionaire is “the rich” in the context of eating them, you’re out of your mind

Everyone's all "eat the rich" until it's one of your favourite. Then it's "No, leave my emotional support millionaire ALONE"

@trash_panda97 if we’re being honest that’s the true Obama successor - she just needs to take office in more of a bush ii and less of an obama economic situation

for all her faults, i think AOC earnestly believes the best hope for the american left is her, as bernie’s successor, being able to stitch together a left-liberal coalition to beat both newsom democrats & trump republicans, and the moves she makes flow from that theory of change

@joni_askola how do none of you actually know hasans takes on this stuff? just constantly making shit up, it’s wild.

A clear example of the horseshoe theory in action: These two people hold very similar views on foreign policy. Both support Russia, China, Iran, and their proxies. One is a far-left streamer, and the other is Trump’s Director of National Intelligence

@charlespacker The answer to the last question has always been explicitly against the TOS?

I don't get how this works. Claude OAuth + OpenClaw (OSS) = banned Claude OAuth + "personal software" using Agent SDK = OK So if I fork OpenClaw to use Agent SDK under the hood, this is OK? Or is it only OK for "personal use", and if I sell it, I am banned from allowing end-users to connect their own Claude OAuth? I think the question people want to know is "can businesses allow end-users to bring-their-own Claude plan, if using the Agent SDK under the hood"?

@jumperz It’s always been against tos my dude, and you’re still welcome to use the API. All of this performative outrage is getting exhausting.

let me get this straight: >you sell a model >someone builds real infrastructure on it >it works so well you can’t handle the load >so you email your entire user base blaming the builder >then offer him the door you didn’t just lose a power user, you just told every dev watching that if they build too hard on claude they’re getting kicked out… aws would never hit up their biggest customer like “hey bro you’re using too much aws lol” but anthropic just did and threw in a coupon..

@ginamilan_ Shouldn’t be hard to provide an example?

Never thought I’d see Americans rooting for Iranians to capture U.S. soldiers. Fucking sick.

@rushicrypto It felt a lot like this during the Bush administration too, tbh.

People who are older than I: Does it really feel like end times or is this just a temporary dip? Because it really feels like end times.

@KevKYan @SyntheticBeef @reactive_dude When I think “scale” I think exabyte distributed storage or the like. And I didn’t mean enterprise plans, just regular API usage. Regardless of LOC, you just need very clean architecture documentation, things to be meaningfully named, and the architecture nicely modularized.

@gnawbone_ @SyntheticBeef @reactive_dude You seem confused. Scale =/= enterprise. Enterprise plans cost 5-10x the price for things like observability, audit trails, compliance, etc. I can be a consumer and still build at scale.

am i stupid for always using opus on max for all coding tasks? does it make sense to use other settings, if i’m not concerned about the rate limit?

@KevKYan @SyntheticBeef @reactive_dude I’m not sure how you’re defining “scale” here, but in that case why are you even using a consumer plan?

@gnawbone_ @SyntheticBeef @reactive_dude It absolutely does when you engineer at scale. If you wanna scope tiny tickets for a single repo, fine. You want to develop at scale and speed, you allow the ai to take in extra context from the codebase to reduce your own manual work.

@KevKYan @SyntheticBeef @reactive_dude Project size has nothing to do with being able to scope tasks.

@SyntheticBeef @gnawbone_ @reactive_dude If your projects aren't big enough to hit limits just say so dude. No need to get all insecure about it. Half the internet's talking about the same problem and u think it's hallucination 🤦‍♂️

@ApexJones22 first, they weren’t trained you nitwit second, the prequels stuff was dumb third, you probably hated the acolyte which easily had the best lightsaber fights of anything

Somebody in the comments said Disney never misses an opportunity to miss an opportunity

@reactive_dude @KevKYan You’d be shocked at how awful people are about managing context/cache.

@KevKYan Which plan do you use? I find the $200 plan more than enough for my needs, and I rarely come even close to my 5-hour window limit.

@AzFlin then add in using dangerously-skip-permissions and walking away, coming back 6 minutes after the last message, and then slamming all of that cold cache

> opus 4.6 > thinking mode set to high > plan mode perma on > 15 diff skill MDs installed on coding best practices "yo guys wtf my claude usage already is already full"

@VinceDaoTV “conduct foreign policy”

“The US can’t conduct foreign policy because my local municipality is corrupt and money laundering”

@Pr0Ger @Pranit tell me you don’t understand how LLMs work without telling me you don’t understand how LLMs work

@Pranit Two prompts? I woke up today and decided to check one thing Meanwhile it’s almost impossible to burn limits on codex

Anthropic is burning customer goodwill over something they already know how to fix They price every API call to the cent but won't tell subscription users what their plan actually buys People can accept tighter limits What they can't accept is being told they all started using Claude wrong on the same day People are reporting 40% of their Pro limit gone in 2 prompts. Full weekly allowances drained in 19 minutes. Single-word messages eating 1-4% of sessions That is not a context window problem. It's a transparency problem Anthropic is precise where it benefits them and vague where it protects them The API pricing page is exact: Opus 4.6 runs $5 per million input tokens and $25 per million output. Every call metered The subscription page says "more usage" and "5x or 20x more usage than Pro" More than what? Anthropic has never published how many tokens each plan gets You can see what percentage of your limit you've used, but you can't see what the limit actually is. And you have no way to verify that what you're getting today is the same as last month When Codex users started hitting rate limits faster than expected, OpenAI reset everyone's limits while they investigated. They said they didn't fully understand why and chose the cautious path Anthropic's response was to tell paying users to downgrade - Switch from Opus to Sonnet - Turn off extended thinking - Cap your context at 200k instead of the 1M they advertise - Start new sessions instead of resuming old ones In short, 'this is a skill issue' Every one of those is a feature people are paying to access Paying users should not have to reverse-engineer what their plan buys If Anthropic wants to rebuild trust, the fix is straightforward - Publish actual token budgets per tier, the same way they already do for the API - Show what each message costs against the budget - Let people verify for themselves whether the deal changed They already do this on the API side. The choice not to do it for subscriptions tells you everything

@Tabbyanne1984 @ryangrim Did your congressional representative approve this war?

@ryangrim This isn’t the time to make jokes Ryan.

Obviously what these downed pilots need to do is find an old, abandoned fighter jet in an unguarded hanger and steal it

@d_h_sanders x.com/catturd2/statu… Lots of gems in this thread

Any updates on this?

@sri_batchu People are insane. They’re recommending to downgrade to Sonnet _on a Pro plan_. OpenAI has it right by gating the top models behind the top plan.

Anthropic is usually very good at handling PR for product issues. But this one is not it. If it was a cost-saving move to massively increase token price and increase dispersion btwn api vs non-api plans. Company should own that instead of trying to gaslight the thousands of users that feel the change viscerally from similar usage. Asking people to just downgrade to inferior products is pedantic and adds insult to injury.

@asmirkn @kimmonismus are you ablooblooing that you only got $3800 of usage out of your subscription?

@kimmonismus They are simply trying to protect their business for themselves. I have proof that there are issues atm, you can run npx cchubber for yourself to find out as well

I don’t get this post. All users are saying they're reaching their rate limits faster than before. Anthropic investigated and said: we found nothing, but hey, why not use Sonnet instead of Opus because Opus uses about twice the bandwidth?

@ImNikhil117 @kimmonismus If you’re on a max 20 plan and hitting limits you’re probably doing something wrong.

@kimmonismus Paying $200/mo specifically for Opus and then being told "just use Sonnet" is wild. That's like a gym charging premium for the squat rack and saying "have you tried the treadmill?" when it's always taken.