> When we do land on something, if it affects existing subscribers you'll get plenty of notice before anything changes. Will hear it from us, not a screenshot on X or Reddit.
If you don't want things like this spreading through screenshots of X and Reddit, don't run "tests" like this in the first place!
(Also "if it affects existing subscribers" is a cop-out, I need to know the pricing of Claude Code for NEW subscribers if I'm going to adopt it at a company with a growing team, or recommend it to other people, write tutorials etc.)
pretty much none of these big providers are offering the guarantees needed to be taken seriously in workplaces right now. the technology itself isn't offering the deterministic guarantees that should warrant it in the workplace right now. problem is everyone's foot is just on the gas. even if your workplace isnt paying for it, people are just straight up rolling their own personal claude accounts to do work at orgs.
ive been trying to make the case all year that if we're going to let employees do shit with ai, lets try claude. in the past like.. 2-3 weeks all that goodwill has basically evaporated.
local inference needs to take off asap because all of these entities actually suck and i wouldn't trust a single sla with anthropic. they are not acting like a serious company right now, this is a joke.
Copilot's per-prompt pricing model is overwhelmingly the best value for money right now, although they more than doubled the price of Opus 4.7 compared to 4.6 and completely removed 4.5 and 4.6, which erodes their lead somewhat. Copilot restricts context much more than CC, but I still find it to be plenty capable. I've occasionally managed to give Copilot/Opus 4.6 a prompt that kept it productive for a full work day for a cost of just ~$0.10.
My company has Github Business, which provides all major models.
Claude runs out of tokens usually in the middle of the month. I can use it with opencode.
Then one extra Claude Pro.
Then also:
Codex business for a test month. gpt-5.4 was excellent.
And since kimi-2.6 I bought their monthly for testing. Works fine. Not as excellent as opus, but usable. Fixes tricky problems by its own.
Not an OpenRouter subscription yet. Those models are either free on opencode anyway, or not good enough.
The free MinMax 2.5 on opencode is pretty usable also.
I know of a very serious business that deployed Max to all of their developers. API pricing, from what I see, can become more expensive than just hiring another dev.
We're also not seeing much difference in real throughput at an agency. Everyone is getting decent results, output wise but it just doesn't seem to change the outcomes that much. There is also a mixed incentive at an agency, because a reduction in hours spent is a reduction in revenue.
It will be interesting to see how it all plays out, but I suspect if cost continues to increase and output only improves incrementally from here, that the cost will be the final decider rather than the competence.
I could see it being a thing we use only sometimes, for some things, but ultimately remain reliant on developers to get the work through the pipeline.
Management seems disconnected from reality. Real employees accumulate tribal knowledge, have an almost infinite context, and don’t keep disabling unit tests because they don’t pass. They don’t really cost money if it’s information workers that build almost all of the modern service industry. It’s management that we should see as a cost center.
That's true but employees offer more than code output, and you still need people operating the "machine" at this stage.
I am interested in how corporate politics evolves in this new environment. Usually all the way up the chain, managers and directors use head count as a measure of power and influence (and compensation). Who's going to pay a director top level pay when all they're doing is funneling requirements to various agents? That seems like a technical role that isn't particularly aligned with the soft skills management excel with either.
Well yes it is expensive, but companies are paying for that. It is far more expensive than the Max and it does go up to or more in some cases compared to the employee salary.
Larger companies are using Claude through AWS Bedrock and are willing to easily pay $5k+ per engineer per month for it.
The thinking appears to be that a model that can do the work of a developer must be worth a significant share of a developer salary. I think this idea is flawed.
Developer salaries are driven up by scarcity - scarcity of developer skills overall and scarcity of developer skills in specific places like California. If AI models destroy the scarcity then the price worth paying for a coding agent will drop dramatically.
Maybe Anthropic can get away with it for a couple of months. But this will not last.
But if e.g. a developer can do 50% more, shouldn't it be worth it to pay up to 50% of developer salary for the product?
So the % is debatable of course. There's cases where an AI agent can save weeks worth of investigation, there's cases where you are mainly blocked due to processes, and many different circumstances. It's up to every company on their own to decide it. But if they decide it's 50%, why shouldn't they spend 50% of salary on it?
Like imagine a large company with thousands of microservices. You need to build a feature, before you had to setup cross timezone team meetings to figure out who owns what, what is happening in each microservice, how it all connects together. But now you can essentially send an AI Agent to scour and prepare all this material for you, which theoretically in this planning could save hours of back and forth meetings.
If 1 hour / 1 eng costs $200, then a 10 people 1h meeting avoided would save $200 x 10 = $2000 alone.
I don't see it as a replacement for dev, it's more of a multiplier.
I believe what GP is saying is that there is a price calculation today, but then if enough devs become unemployed, their salary will go down, making them more competitive by finops calculations, at which point the Ai prices will have to come down as well. Where equilibrium is, no one knows
I think it's an interest hypothesis but I don't think it works out like that. AI prices aren't priced in relation to the work they do, they're priced in relation to tokens (input/output). As long as it's cheaper to use those tokens than it is to pay a dev, then dev salaries will likely fall. Whenever it becomes cheaper to hire a dev than to use AI, a company will likely just hire a dev. But AI prices won't fall just because dev salaries have.
Yeah, I mean I think there's just too much work and I think devs who are effective with AI won't become unemployed, but their productivity will be multiplied. More will be expected of companies in terms of output, so it will be just more output.
>But if e.g. a developer can do 50% more, shouldn't it be worth it to pay up to 50% of developer salary for the product?
That's the upper bound but it's not the market price.
Accounting software (+ hardware) doesn't cost nearly as much as the accountant hours it saves. Accountant salaries are simply not a relevant yardstick for the price that software vendors can charge for accounting software.
Equally, the market price for code generators will not stay anywhere near the price of developer hours it saves. It will be determined by competition.
Because accounting software is cheaper due to competition. In software eng Claude is currently strongest and there's higher costs involved than normal SaaS. There are many fields in which the tools/machinery cost more than the salaries of people.
>Because accounting software is cheaper due to competition. In software eng Claude is currently strongest and there's higher costs involved than normal SaaS.
Yes, competition not salaries determines the margins that software vendors can charge. That's exactly what I'm saying.
My expectation is that competition between coding agents will stay strong and costs for the current level of software engineering performance will fall.
>There are many fields in which the tools/machinery cost more than the salaries of people.
Of course not, but they cost more than the person using them, it's multiplying the productivity of that person, so if AI multiplied enough as well it would make sense.
The question was whether the salaries that would have been paid for the working hours replaced by the machine are a realistic yardstick for the market price of the machine.
I am pretty sure that a hole in the pocket in the order of 50 000 000 USD/month (assuming around 20 000 people using AI in not the smartest or most optimized way possible, therefore burning A LOT of tokens) will be noticeable by even the largest companies.
It is noticeable and even promoted, large companies do pay such sums for the API, like $5k+ per person per month. Not every eng is using AI that much already, but companies are clearly willing to pay those sums.
I just cancelled before seeing this news. i was already pissed about constantly hitting limits on the 20 a month plan and looking for alternatives and this seals the deal. Bye bye!
Yea, I've been fine so far, but something happened with Opus 4.6 and especially 4.7. I was able to do some actual work with a Pro plan before. Now it's just pure anxiety of hitting the limits.
With Sonnet it's a bit better, but I can get the same performance with GPT-5.4.
Now I'm pretty much paying the 20€ for Claude Pro so it can plan/review stuff and then I use pi.dev + GPT-5.4 for the actual work.
I just paid for Pro for the first time 24 hours ago. Its been great, but the limits are crazy. It's nice not dealing with ChatGPTs sycophantic gaslighting, and not having random bugs.
That said, I seem to be caught in that 2% test if I open in a private tab. What nonsense. I wouldn't be paying for Claude if it wasn't for its quality abilities, which necessarily includes Claude Code.
I can easily hit the weekly limit on Claude even on the $200 plan. I have yet to ever hit a rate limit on Codex $100. And the results are almost as good. And don't get me started on Anthropic's extra usage scam.
Not the op, but it’s fairly easy to hit if you automate a kanban and have some stuff you want to get done. All those little “wouldn’t it be great if” tasks that show up after doing a big task become very doable, it just soaks your tokens.
To play devil's advocate, without A/B testing a lot of decisions would be made with insufficient relevant data, and lead to subpar results that affect the many negatively form the road.
counter-point : the companies that are most famous for A/B testing routinely are also the ones with the most notoriously non-existent customer service departments globally, facebook/google/amazon/ebay. Groups that harbor dissatisfied customers by essentially being 'the only show in town.'.
so, what i'm saying is : I think a lot of companies align themselves with the cash first and then measure whether or not the negative image/user impact is manageable .
A lot of decisions made with A/B testing are also made with insufficient relevant data, but it's less obvious since it's easy to think the A/B results cover everything.
> Depends entirely on the stakes and whether personal data is involved
Sure. Let me just A/B test whether or not you'll respond positively or negatively to having your news delivered via push notification or delayed by 10 minutes.
I'm sure you would appreciate being tested on without your consent, just so that I can make an extra quick buck at your expense. Nothing amoral or unethical about it.
What do you think about slow rollouts for new features? Like, we think this new push notification system will be loved but let’s ship to only 1% of users in case there’s a horrible unforeseen consequence like occasional 10min delays? Dashboard goes upside down -> revert then work through logs to figure out what the hell went wrong.
So you're perfectly okay with repeatedly paying for a shit product, getting shat on by the company in the form of being tested for feedback, and "maybe" getting a better product in the future. Mind you, that "better" isn't necessarily better for you but more explicitly better for the company you're paying.
Sounds like someone who doesn't care about being a sheep. Or maybe someone whose salary depends on having sheep.
I think you are making far too wide-sweeping statements. I think most people here probably agree that if Anthropic drops Claude Code from the Pro plan after people have paid with the understanding that it is part of the package, that would be wrong, and they deserve to lose business over it. However, there are plenty of situations where A/B testing is entirely benign, and I would not have any problem with a company doing that testing without getting consent first. Every form of A/B testing is not done just for the gain of the company doing the testing.
> I need to know the pricing of Claude Code for NEW subscribers if I'm going to adopt it at a company with a growing team.
I agree, but can you really use Claude Code on the Pro plan as a full time developer, or professional 'knowledge worker' without hitting the usage limits fairly early in the day anyway?
I'm in the academia, and Claude's performance in my field could be described as a very fast junior grad student. When I use Claude Code, I typically spend a few hours figuring out what needs to be done exactly, and describing it in sufficient detail. Then Claude does it in 30 minutes, while an actual student would need days. And then I spend anything from minutes to days evaluating the results, depending on if it needs to be tested with real data and how much weirdness those tests uncover.
But I also have other work to do beyond guiding the automated grad student. Which means my Claude Code usage rarely exceeds 1–2 hours/week.
I use Pro professionally and didn't hit limits most of the time. I believe I used up 5hr quota once or twice. We switched to Team sub and I'm on Standard(which is Pro x1.25 I believe). I don't vibecode entire applications, I ask it to make boilerplate, smaller, well scoped features or fix some errors. I don't let it go off with a prompt "make another netflix clone" cause I just don't see any real value in that
Just the Pro Plan Claude Code on its own? Maybe you could last a full day on just using Sonnet. Maybe one Opus dab in the morning to plan your Haiku/Sonnet day?
I have Pro Claude, Plus GPT and Pro Gemini. When one runs out I switch to another project on the next LLM. If I really need a task finished I'll restart it on another LLM, but I'm loathe to do that as it eats tokens just getting back up to speed.
It’s pretty reasonable to say “demand is way up, quality is up, supply is constrained, and so price needs to rise”.
It seems weird to segment this way though. Surely it’s better to just give Sonnet to your bottom tier, rather than cut out the entire Claide Code product entirely?
Give folks a taste rather than lock the whole product behind a $100/mo plan.
But if Sonnet is bad it would give bad impression of the product, no? And it also takes compute, so you give a bad hallucinating impression of your product while still losing compute.
It’s not bad though, it’s crazy good in comparison to any model older than 1y old. If you don’t have access to any vibe coding at all it’s gonna be life changing.
But I think you are right, as long as Codex and Gemini are cheap alternatives then vs. 1yo models isn’t the correct comp.
Then it’s probably better to just resegment the whole Claude Code product as an enterprise only tier. (That also has the advantage of kicking out all the Claw subscribers that screw over the token limit economics for normal $20/mo users.)
I mean, this is why they do A/B testing. This way of testing stuff is not new at all, people who act genuinely surprised need to do a reality check. Companies want to maximize profit. They do this by testing what creates the biggest profit. A/B Testing is one of the ways to do this, and it has been used for decades in precisely this way.
Maybe a silly bet where the head of sales had 1-2 glasses of wine too much... "I bet they will still pay us 20 bucks/mo without CC! Don't believe me? I'm going to prove it!"
>"his title should be changed to Head of Corporate Bullshitting"
They're hitting the physical limits of energy production and chip supply for inference capacity. There's literally nothing that can be done but reduce usage to spread it around for now.
Hopefully the negative responses in that thread + the conversation here on HN might help them realize that totally removing Code access for Pro users isn't a good look.
And with no free trial period on top of that, nobody is going to want to pay $100+ just to check it out. I can't imagine the conversion rate of that test being positive.
good reminder to me to stay in practice with manual coding for my side projects! claude is super convenient for them for now but if it goes it goes, I definitely don't want to get dependent on it. maybe the local models will improve in a year too.
CC has such egregious API subsidies that it’s hard to not to leverage it unless the license tells an enterprise otherwise. Love the subsidized pricing while it lasts.
> CC has such egregious API subsidies that it’s hard to not to leverage it unless the license tells an enterprise otherwise.
It's hard to tell, honestly - about half the HN population will tell you that all the token providers are running inference at a profit when using the API and only the subscriptions are subsidised, while the other half will tell you that everything, including both the API and the subscriptions, are subsidised (i.e. running at a loss).
My company currently uses the Anthropic Enterprise subscription plan, but we’ve been informed that’s going away in 2027 in favor of API billing. If businesses are using subscriptions, I don’t think they will for long.
If your definition of "real businesses" is "Fortune 500, US based tech company with more money than sense or just happy to bleed VC money", sure, 99.999% of businesses are not real businesses.
You may also have a very narrow view of how the world actually works, left as an exercise to the reader to figure out which one it is
I don’t get the surprise or discontent. People hooking themselves up to a paid SaaS that only two vendors can offer (Anthropic and OpenAI), no competition or regulation to speak of… of course they’ll do whatever they want with their plans.
Hope you can still resume working on your projects without AI.
> Hope you can still resume working on your projects without AI.
How often does technology really move backwards? I can already run a decent local model on a spec’d out MacBook and for better or worse electronics/computers moves one direction (with minor supply chain hiccups along the way notwithstanding)
If one doesn't want rug-pulls, one signals their policy makers to create regulation to prevent it. Otherwise it's just... uncapped capitalism or what's the name
Yeah I flat out don't believe the 2% thing. It's possible that I was the 1 out of 50 who checked the page and saw that Claude code was removed... but it really seems like everyone I shared it with saw the same thing which is incredibly unlikely. Also I am an existing subscriber and checked the price page while logged in, so I shouldn't be counted in "2% of new subscribers" at all...
Yep, and the price point theyre looking at is 95% of an engineer.
Once they get people hooked, deskilled, and paying, the money ratchet only tightens.
And the companies KNOW that theyre replacing engineers, or trying to. So each engineer replaced is X salary a year they now have available, so make it back in SaaS LLM tokens.
Is it? I’m curious because I thought they were raising prices to pay for exorbitant training costs, not because subscribers are expensive on a unit basis.
I thought inference was cheap so there was little marginal cost of a new subscriber.
It is honestly truly fucking incredible how corps still find new, innovative ways to enshittify. Regular enshittification won't cut it, they have to exercise their artistic creativity. Who the fuck comes up with the idea that what services you get with your subscription are random? It's mind-boggling that some percentage of people visiting the website will be presented with an inferior version of the same subscription for the same price. I'm not even mad (despite my colorful wording), I don't use Claude, just impressed with the bold new territory being explored here.
I think of enshittification as "we're making plenty of money but let's make more." In other words greed.
Based on how much money Zitron has reported that these companies are losing on every subscription, this feels more like they're just trying to survive. In other words "ohshittification."
My take: it is not enshittification to raise the price for a product whose demand outstrips its supply. That is basic economics. There are alternatives, it’s not a monopoly. If you think it’s the best product, then pay more for it.
Personally I would be perfectly content if the price of Max went up a bit and Pro no longer worked for CC if it meant that Max was faster and more stable.
> It is honestly truly fucking incredible how corps still find new, innovative ways to enshittify. Regular enshittification won't cut it, they have to exercise their artistic creativity.
I had a bit of an epiphany the other day thinking about these VC companies offering products to the public at unsustainable prices. It's classic anticompetitive behavior.
You imagine anticompetitive behavior to come from a monopoly because they can afford to burn money to drive competition out before they bring prices back to profitable but the whole VC burn is the same thing. People talk about it a lot without really saying it explicitly when they talk about moats. The only moat Anthropic and OpenAI have is money and they utilize it by offering products below cost.
The two companies are just trying to outlast the other one until they are the only one left.
So it's not really enshitification as much as you were previously getting the deal of a lifetime.
In physical markets we call this kinda thing dumping and it's often regulated. Maybe offering SaaS or compute at below profitable rates should be investigatable too, to avoid killing competitors too easily?
Dumping is typically used in the context of international trade.
There are some predatory pricing laws, but they're much more narrow than most people believe. There is no law requiring things to be sold for more than it costs to produce.
I think it's funny that these topics make people angry enough to demand that we make laws to force companies to raise prices. We'll stick it to these companies by forcing them to charge us more! That will show them!
Such laws would be very bad for startups and newcomers because they'd be forced to price their new product higher than established competitors who have economies of scale. It would be a nice handout to the big companies.
The whole Silicon Valley VC industry and the majority of the net worth of SWEs on HN is based on dumping. "Burning VC cash" is transparently dumping, and it's squarely what the US big tech dominance is founded on. Amazon, Uber, Youtube, now LLMs. The huge majority of "success stories" of the last 15 years are based on dumping their product far below cost price, running at a loss for years until they dominate the market, and then jacking up prices/enshittifying/selling user data.
This happens naturally because no company can run at a loss forever.
I think it's funny that we're getting subsidized and discounted services and this makes some people so angry that the comment section is demanding laws that would force companies to charge us more.
Of course but they can run at that rate long enough to make it impossible for another company to compete and go bankrupt. Which is the problem.
Like I said, this is the iconic strategy of monopolies. They take their pool of money, move into a new market, lower prices below cost until local competition withers away then buy their assets at bankruptcy prices and raise prices to whatever they want. Those temporary discounts are designed to kill competition, innovation and choice.
VC subsidized prices are exactly as bad for the market as that. It's unclear to me if the intent is the same but intent doesn't really matter.
I'd also like to state that I'm not angry about it.
It could be an A/B test to see whether people without an existing subscription care about Claude Code (CC) at all. If they sign up then CC is disabled (or not as it is not really an issue to offer more). Capturing that info would definitely be useful to a growth team.
They need to give the non-code service a different name.
And a slightly lower price.
If it succeeds they can adjust pricing later.
Otherwise they are messing with their new and old customers heads, regarding a service with a name that ought to be reliably interpretable. And seriously messing with their own credibility. Wrong kind of A/B test.
This is incompetence which i would normally discount. But Anthropic seems to be falling all over themselves to irritate customers.
No I think the test is that some new sign ups won't get Claude code in that tier if they pick it and they're seeing if users will still pay for it without it?
Although the ones that never touch claude code are a free $20 a month, the ones that do are potentially a seventy to eighty dollar twenty dollars a month . it’s not instantly obvious which customers you prefer (revenue vs cash negative growth- on second thought obviously they prefer the second)
They've preferred the second so far, but they might have a fair reason to see if they can keep growing with the first one instead or cut down on some loss leading, right?
That's how i read it too - they want to test if people will still pay for pro plan if it doesn't include Claude Code. At the same time they are also saying that if you subscribe having been told it does include Claude Code, they may still change their mind later and take it away!
Random data point: Guest passes apparently still include Claude Code in their Pro trial. If they are running a test this is a really sloppy way to do it.
Apparently it's just an A/B test. Legit LMAO moment, speedrunning reputation destruction to your entire userbase just to test a question whose answer you can probably already guess.
---
> For clarity, we're running a small test on ~2% of new prosumer signups. Existing Pro and Max subscribers aren't affected.
> When we launched Max a year ago, it didn't include Claude Code, Cowork didn't exist, and agents that run for hours weren't a thing. Max was designed for heavy chat usage, that's it.
> Since then, we bundled Claude Code into Max and it took off after Opus 4. Cowork landed. Long-running async agents are now everyday workflows. The way people actually use a Claude subscription has changed fundamentally.
> Engagement per subscriber is way up. We've made small adjustments along the way (weekly caps, tighter limits at peak), but usage has changed a lot and our current plans weren't built for this.
> So we're looking at different options to keep delivering a great experience for users. We don't know exactly what those look like yet - that's what we're testing and getting feedback on right now.
> When we do land on something, if it affects existing subscribers you'll get plenty of notice before anything changes. Will hear it from us, not a screenshot on X or Reddit.
Personally I love how they have increased everyone's quotas to counteract the Opus 4.7 tokenizer change a few days ago, but are immediately regretting it and trying to cut off subscription users.
If the subscriptions are unprofitable, then just communicate honestly, raise the price or lower limits for new subscribers transparently, and grandfather in existing users. That's what GLM coding plan is doing and it works fine for them. Don't ruin your reputation with opaque messaging and hidden changes. Lol
> If the subscriptions are unprofitable, then just communicate honestly, raise the price or lower limits for new subscribers transparently, and grandfather in existing users.
This. Why do so many companies fail to get this? Anthropic's user base, in particular, is intelligent enough to understand their constraints.
It's possible they don't know the actual costs. The one-time costs like hardware and training foundation models is huge. There's the ongoing costs like the PhD dudes they hire. There's the data they buy and decide not to use. There's the various offers under one payment plan - limited artifact hosting, cowork, image upload, Claude Code, Code tab on Claude. What about feature parity between the site and app? Who's working on all these?
tldr it seems really complex and by the time they've counted it they probably hired 40 new people for an unannounced feature.
I mean, if you look at the Claude subreddits, the general consensus is that Anthropic and OpenAI are money hungry corporate devils that are here to enslave all of us, and everyone should get unlimited everything on Max x20.
I think you may be overestimating the willingness of people to understand Anthropic’s concerns.
Reddit is a very small subsection of the Internet consisting mainly of basement-dweller radical leftist they/thems. We should not be taking them as a representative sample of the average Anthropic user.
Ironic that you are talking about unrepresentative samples while characterizing Reddit users that way. Reddit is a huge subsection of the internet, over a billion monthly active users. It covers basically all strata of western society you can think of.
They seem to count unregistered users in the D/W/MAU, but even so the producer / consumer of content ratio across social media still stands. What you read is a subsection of all visitors to Reddit, which follows the power law.
I have to ask - are you a bot? You seem to be making some very strong and misinformed statements here. You can fixate on the number if you like, but calling Reddit homogeneously radical left or 'very small', is pretty stupid.
big fan of A/B tests that dehumanize the consumer into some kind of money making lab rat funnel whose only purpose is to be experimented on how you can extract more money out of it
While I feel the same way, this is nothing new at all. Basically every company does this and it's a totally normal way to test new profit models. Has been done for decades. People acting surprised here really need to get on with reality.
MS paused Copilot subscriptions because they don't have enough capacity. Anthropic is trying to confuse new users and literally don't want 20$/mo because they don't have enough capacity. Seems like there's a trend here. A lot of people in scaling threads were saying that capacity projections and DC buildouts were "fantasy" a few years ago. Not so much anymore...
Usually A/B testing is just on the surface, and when you actually subscribe you get the "better" terms of the possible options.
Like, they're just advertising different terms to test how many people would still click on it and very likely start the subscription process, but after they click they go back to the usual terms. Changing the whole payment flow, account models and permissions in their backend just for a quick test is usually too much work.
But yes, basically, if you're B and not A, and B has objectively worse terms than A, then you're just unlucky. But this is the essence of A/B tests. They are done by basically every company everytime, because it's the most straightforward and simple way to test new terms or designs.
I don't think anyone is questioning the idea of A/B testing in general, but we are confused how it works in practice when you are advertising a service. Can you give me an example of this happening elsewhere? Or are you saying it's so normalized that there would be no record or news articles about it, and this only is newsworthy because "AI"?
I have never heard of a company intentionally advertising less features to only some of their customers to see if they'll still bite.
And we still have no actual confirmation that Claude is giving people who fall into that 2% of testers the full service, as opposed to the advertised one.
I don't think I've ever been on such a rollercoaster with a company's reputation in the developer space. I started in January on the $20 plan, essentially my first agentic AI programming. I quickly started hitting limits developing several apps at the same time. I went up to the $200 plan after seeing the value.
After seeing my own issues with 4.6 and the mega-post on Github about declining metrics in a decent dataset of claude chats by Stella Laurenzo at AMD (https://github.com/anthropics/claude-code/issues/42796), I downgraded to the $100 plan. Hallucinations. Laziness. Lack of thinking. The responses on those mega-threads from Anthropic rubbed me the wrong way in a "you're holding it wrong" kinda way.
In the past week, I downgraded back to the $20 plan because the Codex $20 plan on 5.4 was working so well for me.
Then throw in other oddball events like the source code leak, and the super positive Anthropic events like their interactions with the current administration. It's a wild ride.
I can't understand removing Claude Code from $20. I'm interested to see whether this is confirmed or not.
I'm a career engineer and I went from being one of their most outspoken proponents (at least within my circle) and now.... I'm not.
Same loved them, told my team about them, got them to switch off of cursor, now I'm telling them to swap to Codex.
Anthropic really pissed me off with their harness crap. They're well within their rights but their communication over it was enough to get me to swap. I don't need extra hurdles when there's a perfectly valid alternative right there. They don't have the advantage they think they do.
I think we are inevitably heading to using the cheap Chinese models like Kimi, GLM, and Minimax for the bulk of engineering tasks. Within 3-6 months they will be at Opus 4.6 level.
This was literally my task today, to try out Qwen 9B locally on my, albeit a bit memory-constrained at 18GB, macbook with pi or opencode. Before reading this update.
MiniMax has its own issues. Server overloads, API errors, and failure to adhere to even the system prompt. It can happily work for hours and get no job done.
I ran OpenCode + GLM-5.1 for three weeks during my vacation. It’s okay. It thinks a lot more to get to a similar result as Claude. So it’s slower. It’s congested during peak hours. It has quirks as the context gets close to full.
But if you’re stuck with no better model, it’s better than local models and no models.
I have to say, OpenCode’s OpenUI has taught me what modern TUIs can be like. Claude’s TUI feels more like it’s been grown than designed. I’m playing around with TUI widgets trying to recreate and improve that experience
> I have to say, OpenCode’s OpenUI has taught me what modern TUIs can be like. Claude’s TUI feels more like it’s been grown than designed.
Claude's TUI is not a TUI. It's the most WTF thing ever: the TUI is actually a GUI. A headless browser shipped the TUI that, in real-time, renders the entire screen, scrolls to the bottom, and converts that to text mode. There are several serious issues and I'll mention two that do utterly piss me off...
1. Insane "jumping" around where the text "scrolls back" then scrolls back down to your prompt: at this point, seen the crazy hack that TUI is, if you tell me the text jumping around in the TUI is because they're simulating mouse clicks on the scrollbar I would't be surprised. If I'm not mistaken we've seen people "fixing" this by patching other programs (tmux ?).
2. What you see in the TUI is not the output of the model. That is, to me, the most insane of it all. They're literally changing characters between their headlessly rendered GUI and the TUI.
> Claude’s TUI feels more like it’s been grown than designed.
"grown" or "hacked" are way too nice words for the monstrosity that Claude's TUI is.
Codex is described as a: "Lightweight coding agent that runs in your terminal". It's 95%+ Rust code. I wonder if the "lightweight" is a stab at the monstrosity that Claude's TUI is.
From this morning: I had a single go file with like 100 loc, I asked it to add debug prints, it thought for 5+ minutes, generating ~1m output token and did not actually update my file.
Anthropic will kick and scream as those are often distilled from their latest models and is cutting into their margin. Though it is not like their hands are clean neither, it is just a different type of stealing, an approved one :-)
Getting them running is easy (check out LMstudio or ask one for some recommendations). The real question is whether you have the hardware to make them run fast enough to be useful.
This is possibly a hot take but recently I've been having about as much luck with Composer 2 in Cursor as I have with Opus 4.6 in Claude Code.
Opus is obviously the better model, but Cursor's "harness" is doing so much heavy lifting in terms of just magically supplying the broader context the model needs to understand the ramifications of its edits.
One thing I enjoy about Cursor and Codex mac apps is the embedded preview window. I know it's not as hardcore as the terminal/tmux but it's hella convenient. But Cursor bugs me with the opacity around what model I'm using. It seems deliberately to be routing requests based on its perceived complexity. What draws you to codex vs cursor?
I think removing Claude Code from the $20 tier is a terrible idea, I never would've gone from nothing right into the $100/200 tier. The $20 plan let me get my feet wet and see how good it could be, and in less than a week I was on the $100 plan.
I think they need to at least have a 1 month introductory rate for the max plan at $20, or devs that decide to try out agentic coding just won't go to Anthropic.
That leads to downstream impacts, like when a company is deciding which AI coding tools to provide and the feedback management hears everyone is already used to (e.x.) Codex, then Anthropic starts losing the enterprise side of things.
I suspect a lot of people are like me. They got into this at the $20/month level individually to check things out. I'm not stressing things out, so I haven't moved up, but the moment I bump into a limit, I'll pull the trigger by default. Until then, I'm the sleeping dog, and you should let me lie.
Well, Anthropic decided to kick me. Now, I'm investing the time to figure out how to use the "open" and "Chinese" models assuming that Anthropic is about to screw me. Once I switch, Anthropic is going to have to demonstrate significant improvements over what I'm now using to get me to even consider them again.
I know of 4 companies that are already starting to stomp down on the AI whales using the "$1000 per day". There was the cost of the entire AI usage of the company and then there was the cost of about a half-dozen people who dwarfed it individually.
So, we've established a hard upper ceiling for what AI can extract per user at roughly $100K and more realistically at $10K per year. Basically, if using the AI costs the same as a human salary, it's going to get pushback. I mean, the whole point was to get rid of those pesky human salaries, after all.
So, there are about 2 million-ish software jobs in the US? It's more than 1 million but a far cry from 10 million. So that pencils in at $20 billion in the US per year total? That means that if an AI company literally won all the US software programmers, it would be worth max $200 billion to be bought out (10x revenue).
Now how much investment have the AI companies taken? Yeah, roughly that. And investors are going to want quite a bit more than that back.
Even if they had zero delivery costs, the AI companies are cooked long term. The moment your number bumps into "All the X in the US/World", you've got a problem.
Short term? Greater fool theory applies. And there appear to be a lot of them.
And all this is before we start getting into people exploring the open models. Most people were like me; we started on something like Claude and just stayed put because it was straightforward. Now that we've been kicked, we'll start looking at the other options.
LLM monsters are deeply unprofitable, going by the industry hearsay (which is the only thing we have, given ultra secrecy of the LLM corporations). The only two LLM companies which disclosed their finances without lies, were two Chinese corporations and they, unsurprisingly, were deeply in red.
Remember the old saying about boiling a frog? LLM corporations need to make most of their users pay hundreds per month, asap. This is Anthropic increasing temperature regulator under the pot just a tiny little bit. Not the first and not the last time.
Tax breaks, fee reductions/waivers, direct monetary incentives, and shielding from "unfavorable" regulation -whether local, state, or national- are all subsidies. Hell, depending on the particulars, government contracts can be subsidies... there's more than one government engineering project out there that could be reasonably referred to as a "jobs program for PhDs", and still more that are corporate handouts.
Every business believed by its "home" government to be sufficiently important gets subsidies when it asks for them... regardless of what nation houses that government. If your claim is that the major players in the "AI" industry aren't getting subsidies from local, state, and national governments in the US, then my claim is that you are lying.
Matches my experience very well. All the goodwill earned from taking a stand against the DoD seemingly forgotten in a month. Coincidentally, I canceled my pro subscription and got set up with OpenCode and OpenRouter last night.
I just installed it and continued about my business. I don't have a carefully tuned claude code setup. It has some skills, some CLAUDE.md files, and not much else.
> I can't understand removing Claude Code from $20
Not according to their webpage: "Claude Code is included in your Pro plan. Perfect for short coding sprints in small codebases with access to both Sonnet 4.6 and Opus 4.7." [1]
There are clear contradictions across their marketing site. As others have pointed out, it's being removed from some help articles and the pricing chart now shows it revoked. Confusing signals, but they seem to be changing all pages in this direction and haven't updated that one yet.
I've noticed that there are some people who feel that their claude instance could be working on churning out multiple apps, and therefore if it isn't they are in some sense falling behind. it's the illusion of productivity raised to the level of a minor addiction.
I had a similar ride, but disagree with your conclusion. Opus 4.7 is so incredibly powerful from my experience, that nothing else really matters and I think at Anthropic they know it. People will pay a lot for access to this model.
Yea, I've seen a lot of whining online, because its more expensive, but from the interactions I've had I'd say, that it's well worth it. To me it feels like another step change, similar to when 4.5 was introduced. Definitely a different beast.
EDIT: it is also surprising to me that everyone seems to believe the people at Anthropic are simply incompetent and recklessly risking their good reputation, while very few consider the possible good reasons they might have for taking such drastic measures. And I don't think it's because of financial pressures in their case
I can’t say I’ve used it extensively enough to draw a conclusion, but it did seem similar to GPT 5.4 in Codex.
When I threw it at a difficult issue in an iOS app, it like GPT came up with wrongly guessed explanations. It only found the issue after I had it instrument the app and add extensive logs. Usually GPT 5.4 is the same.
Only that with GPT 5.4 it’s at least included in my subscription, while sending 3-4 messages to Opus 4.7 for this blew through my $20 plan limits and consumed $10 of extra usage on top. At that point I can’t help but bring up how much more expensive it is.
> Only that with GPT 5.4 it’s at least included in my subscription, while sending 3-4 messages to Opus 4.7 for this blew through my $20 plan limits and consumed $10 of extra usage on top. At that point I can’t help but bring up how much more expensive it is.
Rest assured OpenAI won’t want to leave that kind of money on the table…
Having too much usage and not enough GPUs is a form of financial pressure, no? since you want to replace your less profitable customers with more profitable customers
I've had completely the opposite experience. I've asked for it to research things and it's just told me to "paste xyz into google". Just now I revisited a chat that's 5 days old and asked it to check again (because what I was looking for might have changed), and it said "no".
It's funny how experiences can be so different. I wonder if this comes down to context. My interactions so far were fairly high-level and in some cases it having a strong opinion was actually super beneficial to the outcome. To me it seemed opinionated, but in a very good way. I can see how this could backfire though and have heard similar reports.
Incredible, powerful, but I couldn't believe how fast I hit the limits compared to how it was with Opus 4.6. They removed Opus 4.6 completely from CC. I would prefer it with the previous limits.
That's not how you keep your customers. None of these agents have a moat, I moved away from Cursor when they started doing what Anthropic is doing now, and never went back even when I was a paying customer since the start.
Opus 4.7 may be incredible but for how long? And they may have Mythos but I feel like they will put it out if pressed too much by their competitors. And again for how long will they keep the advantage?
At the speed everything is advancing I don’t think it’s such an advantage. They all catch each other up pretty fast. That’s why I prefer to pay Cursors and have access to all of them instead of being lock to a single one (even if that means to lose some discounted credits). If they opened Mythos today at a good price that would be something but that’s not the case and it won’t happen.
they need the devs on board for that to matter, i can get whatever i want done with lesser models already. It is quite literally about just who is not gonna give me the shittiest experience, and at anthropic it sure seems they are determined to annoy everyone since they started gaining in popularity.
FWIW their "Fin" agent on their web page tells me:
"Claude Code is not included in the standard Pro plan. It's only available with premium seats on Team and Enterprise plans.
However, Pro plans do include improved coding capabilities through our GitHub integration, which lets you connect code repositories directly to Claude. You also get access to code execution and file creation features in the Claude apps.
For the full Claude Code terminal experience, you'd need to upgrade to a Team or Enterprise plan with premium seats"
Yes, confirmed directly from Anthropic's website. Claude Code is included in the Pro plan, described as perfect for short coding sprints in small codebases with access to both Sonnet 4.6 and Opus 4.7.
Claude
You said: Is claude code included with the pro subscription
Is claude code included with the pro subscription
3:11 PM
Claude responded: Yes, Claude Code is included with the Pro plan ($20/month).
Yes, Claude Code is included with the Pro plan ($20/month). You can use it in the terminal, desktop app, VS Code, JetBrains, and on the web.
The main caveat is that Claude Code draws from the same usage limits as your regular Claude chat — so heavy coding sessions can eat into your quota faster. If you find yourself hitting limits often, the Max plans ($100–$200/month) offer significantly more headroom.
The $20/month, which is now $17, should give you a hint? Models have a knowledge cut-off and will not reflect up to date information unless you trigger a web search.
Do you understand how LLM's work and that they are always behind in their knowledge? Unless Claude does a network call to check its own website, it will give you outdated information. Its a prediction model, its not magic.
What a way to ruin goodwill with the very community they are trying to court. I am on a Pro subscription to use with Claude Code, but it sounds like the days of using it are numbered. I guess I will be trying the latest offering from OpenAI and Google tomorrow and if they are satisfactory I might just switch. Moreover, I have been recommending Anthropic's API solutions up to now to friends and clients. Based on this dumb move I will be now starting with this anecdote and then giving a very hedged recommendation.
Realistically the future of all this is that open models become good enough that LLM as a service becomes a commodity with a race to the bottom in terms of cost. Given where we are today I can easily see open weight models in 2-3 years making Anthropic and OpenAI irrelevant for everyday development work (I justify this like so: if my coding agent is 10x smarter than I am, how would I understand if it did all the right things? I want someone of roughly my intelligence for coding. I can see use cases for like independent pharma work or some such where supergenius level intelligence is justified, but for coding ability for mere mortals to reason about the code is probably more important).
The valuation is obviously based on the premise of their capturing the white collar economy. OpenAI's charter says so openly. And Chinese robots will come for blue workers next.
The economy, not the workers :) It feels like pretty soon white collar workers will be in a “You have nothing to lose but your chains” situation. Except we are not as fit as the proletariat of the past.
In my experience, Codex is better than Claude Code in every way and GPT-5.4 is on par or better than Opus 4.6 at every coding task I ask of it.
You're really not going to miss CC. And OpenAI actually had some foresight to invest massively in compute so they don't run into usage and rate limits like Anthropic does constantly. I couldn't even use CC for more than a couple complex tasks before I was out of extra usage or session usage. It was a maddening productivity killer and I just switched to Codex full time.
I am on Google's $20/month plan, and I usually get about three half-hour coding sessions a week with AntiGravity using the Claude models. The limit using Gemini Pro models is much higher. I am retired so Google's $20 plan is sufficient for me, but I understand that people who are still working would need higher limits.
I am also on a $10/month plan with Nous Research for supplying open models for their open source Hermes Agent. I run Hermes inside a container, on a dedicated VPS as a coding agent for complex tasks and so far I find the $10/month plan is enough for about five to ten major tasks a month. I think it is also a good deal.
If I could get the equivalent of GPT-4 running locally, that would cover like 95% of what I need an LLM for. Tweak this dockerfile, gimme a bash script. I guess the context probably isn’t sufficient for the agent stuff, but I’m sure more context-efficient harnesses will be coming down the line
I have an old Mac Mini with 32G of integrated RAM, and the following works for me for small local code changes:
ollama launch claude --model qwen3.6:35b-a3b-nvfp4
In addition to not having an integrated web search tool, one drawback is that it runs more slowly than using cloud servers. I find myself asking for a code or documentation change, and then spending two minutes on my deck getting fresh air waiting for a slower response. When using a fast cloud service I can be a coding slave, glued to my computer. Still, I like running local when I can!
It would signal quite a fundamental pivot if their "Pro" plan excludes coding but supports personal productivity (Cowork). Quite surprising given most people attribute Anthropic's success to their elevation of coding above everything else. To have casual users locked out of that would be a major hit you would think.
Makes me curious about the internal thinking. One theory being they are in a capacity crisis and knocking Pro users off Claude Code is an emergency brake getting pulled. But an opposite theory is it's a revenue move and they think they have the lock in to pull it off. Especially if they are building up to IPO.
Interestingly the Team subscription which is still $20/month/seat still includes Claude Code. But you need minimum 5 seats. So it could be a way to force people off individual plans and into enterprise plans where possibly things scale better for them, especially IPO/wise. When one user wants it in a company, probably they go buy 5 seats.
I have to assume they're compute constrained and thus need to either raise prices or cut their lowest-margin products (which amounts to more or less the same thing, but with different optics), or turn away new users.
My assumption is that people are able to very easily saturate Pro with Claude Code and therefore even though the quotas are lower (more than proportionally) the utilization of those quotas is higher enough that Pro is less profitable.
I think there is a definite possibility that they aren't compute constrained, but rather trying to improve a sorry cash flow situation before IPO.
Of course, I don't have real insight into available compute, but the vibe slope seems to have dropped a bit, at the same time as new GPUs are being shoved into datacenters as fast as possible.
Their enterprise API customers are literally competing to see who can throw the most money at Anthropic. Anthropic has very little reason to focus on a $20/month user, and with their current momentum (especially since enterprise deals are long-lived) they could remove Claude Code from the Pro plan without any revenue hit. In fact, it may be a huge revenue boost given the strength of the Anthropic brand.
I just switched from the $10 Copilot subscription to a $20 Claude subscription to get general AI and coding in one bill. I guess I'll try out GPT Codex.
gpt allows you to wire their models into other CLI tools, I'm advising everyone I know to lean that direction. Not trying to become hostage to something like claude's ecosystem for the rest of my development career.
Their supply of being able to serve the models is hardcore rationed by their ability to scale up datacenters and GPUs.
They really don't need to subsidize unprofitable customers at this point when there is a line out the door to pay thousands of dollars a month per user, that are revolting because they aren't actually being able to get reliable uptime.
They have all the growth they need for now, they really don't need the cheap users.
Why would you even want a Claude subscription if not for Claude Code? Anthropic is obviously the best for programming, but probably nowhere else. Seems like a good way to onboard people to the Claude Code experience...everyone who's working seriously with it needs Opus, anyway. But, maybe that's the rub, if the Pro plan includes no Opus usage (which I think has always been the case), you might have a worse impression of Claude Code. Codex 5.4 is better than Sonnet, but not better than Opus.
I dunno, I'm no business genius, but I think we're starting to see these companies try to find ways to make money instead of losing it.
The pro plan does include Opus usage. I've noticed the limits on the web client are a bit higher than through CC, but probably more because of the increased token usage of agentic coding in general.
Claude web is actually pretty good for dealing with random projects outside of code. I have a Home Assistant MCP server [1] behind a Cloudflare tunnel exposed to it that makes maintaining automations a lot easier.
On LMArena, Claude Opus is ranked as the best at everything except image and video generation, which it does not support. That may be inaccurate, but it's plausible
People subscribed to chatgpt before there was codex. Why wouldnt a Claude subscription stand on its own without Claude Code? In fact it’s probably a smart move for Anthropic to split it out.
Because Claude Code provides me more value than simply a chat agent. Claude Code is the only thing I pay for. I can do chat things with the free tier across all the LLMs.
I have been using https://claude.ai and, initially, it was good, but, unfortunately, it keeps getting worse. I had it search for contact information for a certain public entity, and in Claude's response, all emails were being replaced with [email protected] or something like that. They also added an absolutely horrendous automatic markdown in the text input, so now you can't even properly enter your prompt. It actively gets in my way and prevents me from typing what I want. Fuck you Anthropic.
They already effectively halved it with the introduction of Opus 4.7 and the new tokenizer that basically gives you about half as much usage for the same price.. Convenient to price based on tokens, and leave what a token is a moving target..
I have a Claude Pro tier subscription; Claude Code, as of right now, is still functional for me. If Anthropic does boot Pro-tier users off Claude Code, I will be cancelling my subscription.
They would probably grandfather existing users in for at least a year or something, you have to imagine. Even if this "test" goes very well and points to removal
This test makes perfect sense with their actions the last few weeks, they think they've done enough to transition into the general public and away from devs and our goodwill no longer is something they should be concerned with.
Its funny that openai, who in my eyes went for the general public rather than devs initially, seems to be semi pivoting and catching all the fallout from anthropic's recent behavior.
It is a massive bummer, up until those few weeks ago, i was hard pulling for anthropic for quite some time, now i just dont care and hope something dope emerges quickly that signals i wont ever have to consider either of them.
I can't find an announcement yet, however the pricing page now shows it's not included, and various support articles have removed any mention of the the Pro plan including access to claude code.
See [1] and [2] for an example of a support article that's had claude code removed as a Pro feature.
I guess this is the beginning of the end for subsidised model access, at least from Anthropic.
Because 'claude -p' is a backdoor for any third party client to use subsidized token pricing, and they've flipped course again to decide they do want the OpenClaw type users as long as they're on the $100 plan.
It was obvious, which is why their earlier decision to attempt to restrict 3p access was hugely unpopular. They're now trying to walk that back but on the condition that those users need to be on the higher plan.
Now they're hurting their popularity with those who actually don't mind using Claude Code. I've been quietly swallowing a lot of performance degradation over the past few weeks as I get the resource crunch, but I'm definitely not going back to copy-pasting between browser and editor. And I have no intention of upgrading to Max or doing per token usage.
The question isn't whether anyone could have missed it, but whether Claude Code has release gates that allow people to require that obvious problems should be resolved. From their release velocity it's pretty clear that they do not.
My usage of Claude Code in the pro plan is definitely metered. Every couple hours I have to wait an hour or two and the last few weeks I've hit my weekly limit on Wednesday.
This is a risky move. I might have paid $20/month for my personal projects but the Max subscription is a bit steep.
Now I'm going to learn more about local models. I don't need to be as good as a frontier model. Good enough and free from all this drama is a win for me
Same, I'm currently unemployed and the $20 help me to initiate many small projects. The recent taxing in the token made me start testing local models on my machine. Tho, Claude works better for the front end part imo.
I've been running pi.dev + codex (GPT-5.4) for a long time for my workhorse stuff.
Actually tried caveman mode yesterday and it made everything SO MUCH BETTER. GPT-5.4 has a habit of being extremely verbose to a ridiculous degree, it's like it's writing a report for a CTO or something and padding everything as much as possible to sound smart.
With caveman it just gives me lists of stuff in a compact format. Perfect.
let's be clear(er) here that you like caveman's format and output, so that's one value of "better". I've seen at least a few (n < 10, so there's that) tests on token use and for THAT value of "better" it's not much different.
"free of this drama" and free is great option for companies, of course most use API billing but let's not forget that there are places that budget is limited and being good enough is just perfect.
I would love someone to play devil's advocate against this perspective:
While these tools stand to enable the democratization of productive capability in software engineering and other tasks (creating a renaissance for solopreneurs, let's say), what seems more likely to actually happen is that entrenched capital will become the only player with real access to this "knowledge as a utility" (was it Altman who called it that?).
We already see this playing out in two fronts: 1) the gradual reduction of services and 2) the DRAM market, where local-first tools (i.e., potential disruptors of the emerging "knowledge monopoly" created by the big AI firms) are being stifled by supply shortages. How many promising small-to-medium-sized competitors are being snuffed out of existence (or never starting) due to the insanity of the DRAM/storage/CPU (soon) markets?
The currently-subsidized access that we have to the big Opus-like models will, in parallel, be gradually be taken away until only the big players can afford it. And in the end what we will have is hyper-productive skeleton crews at a few consolidated firms performing (or selling expensive access to) basically all of the knowledge labor for society, with very little potential for disruption due to the hardware and "knowledge" scarcity engineered (in part, maybe) by this monopoly.
Not necessarily a closely held belief – just a hunch – which is why I want to see what parts of the picture I might be missing.
Devils advocate here - pro and max tier customers for all the major inference providers are loss leaders from the data we have been able to figure out, and reverse engineer. They are effectively a marketing exercise.
The real profitability is selling tokens to enterprise, and enterprise demand is growing so fast that they are short on the total amount of tokens they can generate per minute, and are prioritising rationally - enterprise gets a better experience - instead of optimizing for their lowest paying (and most loss leading) customers.
We are in a hardware crunch right now but that won't be forever, and eventually (likely 2028) we will get experiences like we got in January from pro-sumer accounts again.
Not only because of cost. Mythos has only been released to some of the big tech players because it's "too dangerous" [0] for us little people.
It's easy to see this becoming a permanent position; the latest models and smarts are reserved for establishment members only, the riff-raff get the cast-offs. So the establishment is preserved and the status quo protected.
[0] I'm putting scare/irony quotes around this, but if the reporting is accurate, there is something to this; we built the internet on string and duct tape, it's not hard to see how a very smart AI could cut it to ribbons.
In periods of massive inflation, only the most wealthy survive.
But there's competition out there -- the open-source chinese models. In their current form, I assume that will turn off many people but new models -- based on those -- are likely to appear. Also, OAI and Google will release new models and pick up the lost customers.
With GLM and Kimi getting better and better, with both still providing low-cost coding plans with higher quotas, and with how trivial it is to switch to them even within Claude CLI, I'm not sure what makes Anthropic think their users would rather pay 5x than switch to the competition.
Never mind GLM and Kimi, even GPT 5.4 offers a great plan for $20/mo. Even if it gets increased rate limits after May, it's still quite likely that casual users won't be hitting them on a regular basis.
I signed up for a 3 month plan with Z.ai so I could try out GLM-5.1. That was a few weeks ago when it was still $27 for 3 months, now it's $48 for 3 months. I hit limits at least as much as I do with Claude. I hit a weekly limit at one point and it said I wasn't going be able to get access again for 6 days so I must've somehow hit their weekly limit on day one of the week. And that after several timeouts.
Why is management at Anthropic trying so hard to ruin their reputation with developers? I missed the OpenClaw hype but it was something that kept me excited about my yearly subscription.
It makes no sense to do one of the higher tier plans unless they are directly generating you money.
They care about developers from companies that are on their team/enterprise plans or using bedrock.
Individual users barely matter. That's probably also the same group that decides to switch to Codex/Kimi/[whatever the hottest agent on any given day] on a whim, which Anthropic doesn't necessarily want to do business with.
feel like its beyond optimistic on their part, just starting to hear their name be blended with companies desires on job listings, and they are destroying the goodwill of the devs who surely are the main reason their name has landed there. They aren't dug in like a microsoft, maybe they get some staying power for nocode people who feel trapped, but im done with their nonsense already and won't recommend them anywhere. Other stuff is good enough already to match.
A fraction of them do. Many just use whatever the employer provides to get their job done. HN users only represent a small sample of the overall software developers which is nowhere nearly enthusiastic about new things.
In the end, companies are made of individuals. In my previous company, I'm one of the individual who advocate for Claude Code adoption. Amongst my circle too, most who have the authority to make an impact on the AI direction. Safe to say I no longer suggest Anthropic to anyone anymore.
At my company, devs were the ones pushing for the Claude subscription. Left to management, we would have only had GitHub Copilot – we already have an existing relationship with them and the tool is good enough.
If Anthropic is intent on losing the goodwill of the devs, they might not be happy with the consequences. Their product is quite commoditized at this point – the latest GPT, Gemini or GLM is just as good for most enterprise tasks.
Exactly the same for me, and I'm now ... "worried" isn't quite right, but you know what I mean, that they will back out. But to what? We had copilot before which was cheaper, and worked reasonably well with the A\ models, but I'm not even sure those will be there (Opus is no longer on my cheapest paid Copilot sub at home), and I've no doubt OpenAI will jack up their prices soon enough and/or do the "exclusivity" thing so they can only be accessed by their own clients.
> When we launched Max a year ago, it didn't include Claude Code, Cowork didn't exist, and agents that run for hours weren't a thing. Max was designed for heavy chat usage, that's it.
Is there a wager that this is 100% foreshadowing Claude Code will be removed from the $100-200/month Max plans soon and go to something like API-only? Or only available on like a new $500-1,000/month plan? Restrict the $100-200/month ones to Claude.ai (website) or Claude desktop app only?
Either way, doesn't seem good to say it's a small test and then start justifying it in this direction.
Do they have a substantial userbase for this outside of claude code? The only two use cases for LLMs that seem to have significant traction are programming, and erotic roleplay lol. If they stop catering to devs, who is their market?
FWIW, I just heard this guy (Head of Growth) on Lenny's podcast a few days ago and one of the things he explicitly mentioned was creating intentional friction for growth. This seems to be one of those tests.
It is over for the little guy - home enthusiasts and vibe coders. Too many of them saturating resources for Max users.
IF you cannot afford few hundred dollars subscription go out and breathe fresh air. But if you can, watch where the ball is rolling - few thousand dollars subscriptions and even less programmers.
Agree about psychological impact outpacing likely actual impact, but that’s a relatively temporary phenomena as we are all adapting to the new way things work.
Productivity wise employment is far more than code production productivity in a vacuum, and productivity gains are rarely captured by employees (see famous chart on worker productivity where that correlation changed around 1970). I wouldn’t expect to see much in the next 1-2 years besides noticing effective teams increasing velocity of features.
I think people in forums like complaining about things and aren’t representative of the broader set of people who are just using the tools, so no real paradox. For vast majority of tech jobs, $200/mo is still an absolute steal in terms of what these tools offer. Only the dullest of companies would not realize this.
Fwiw in the 80s-90s computers also didn’t really register in productivity metrics. Qualitative changes occur long before accurate measurement catches up.
Because most people work for someone else and don't decide their own salaries. It's not doubling productivity, but even a 10-20% boost to productivity for a team of engineers means that, as a business, even $1k per month per seat is perfectly acceptable. For consumers and hobbyists that basically kills access.
> Best forget about using Claude Opus models in Copilot.
I noticed this morning that Opus isn't even one of the models in the `/model` command in Copilot. Highest I can get (on the paid, but least expensive) tier is Sonnet 4.6. I'm pretty sure Opus was allowed recently, but not now.
Truly makes no sense. I pay for the $200/month plan and end up using about $3k/month worth of API costs. I imagine that the only reason they haven’t cut me off is because my habits serve as good training data for them.
Guess they’ve decided to move in the direction of allocating compute primarily to power users and enterprise.
But power users are not a sticky customer base. I just bought the ChatGPT Pro plan and would immediately switch over if the model performance is better and/or I get more compute.
Most commercial websites, especially tech, run multiple A/B experiments in parallel to optimize signup funnels and tier conversions. Even so, there should always be a source of truth for people. We should have some way of verifying what we get for what we pay. There are laws about this as well.
This guy's casual and crass response is a sign of disrespect for customers. Unfortunately, that is pervasive in the industry. The bubbles these teams work in are corrosive to empathy and real world impact.
I think there is more here: Anthropic's whole market positioning is based on trust. It's literally their reason for being.
The Claude constitution has a major section about not being deceptive. Now this is GTM, not the model, but there is clearly a coherence problem here... and if anyone should realize the important of their market positioning it's GTM.
Agreed that it’s 100% marketing. In some ways Anthropic is more of a for-profit corporation than OpenAI which is at least partially owned by a non-profit.
Claude has become practically unusable for Pro users in the past few days. The Opus 4.7 blew through an entire 5 hour limit in one question and didn’t even finish answering it. Zero value delivered.
Opus 4.6 is giving 2, maybe 3 questions before blowing through the Pro 5 hour limit as well. We are forced to use Sonnet which makes the same mistakes over and over and then to start trying with other companies. To make matters worse, it reuses old code as we try to survive between credit expiry so it re-introduced issues into the code with the limited credits, that we had already fixed on our own and with other models.
Anthropic in just a few days has gotten me to try GLM 5.1, the new Kimi, and back to OpenAI. OpenAI also seems to introduce new bugs without being carefully micromanaged. The advantage Claude has is that the models are more careful and can refactor code instead of leading to bloat as they go. But the throttling happening now is breaking things and making the entire subscription unusable. I really hope they fix it soon.
I'm starting to think I've been A/B tested, because this was my experience for almost a year with Claude ever since I tried it for coding. Meanwhile, my coworkers seemed to be able to use it for long periods of time without getting rate limited.
One interesting variable is that I'm located in Vietnam while my coworkers are located in Norway and Europe.
To work around this issue I used Claude for coding with a Copilot subscription which was much cheaper and had virtually no rate limiting.
Copilot gives you some set amount of credits each month, but you can also pay as you go if you run out of credit which is much better than the 5 hour window crap claude code would give me.
The only opus model available now on copilot for some reason is 4.7 and it costs 7.5x tokens, while everything else is 1x, 0.33x or free.
But I switched to using GPT 5.4 medium for a month or so which I find very reasonable.
My personal LLM coding stack is now OpenCode, Claude Sonnet for ideation on spec with OpenWhispr for voice-to-text, GLM-5.1 for the orchestrating loop, GLM-4.7 for coding, and DeepSeek R1 for review and validation. It works much, much better than the Claude Code setup I have at work for substantially less money to boot.
At this rate I fully anticipate being able to run a comparable stack on a 128GB Mac Studio using quants of newer-generation distilled OSS models in a year or two. Being able to ramble to a computer for an hour about features and technical philosophy then have it build a nearly-working app for $50 is an exciting feeling. There's still a long tail of productionization and fixing what the model didn't adhere to but it's still incredible.
Im locked in for a year of claude pro, I encountered the same issues as you a couple weeks ago, Id get like one solid plan done and really really hope it was a 1 shot because that was legit all i was gonna get out of it for those 5 hours, and it would be ~10% of weekly usage to really make me feel scared to hit send.
I got the 20$ gpt tier, and now i just use claude to craft MD plan docs instead, and then i hand them off to gpt 5.4 and it has been working great. can do about 4x as much work or so based on my feelings(not accurate). if i have just small simple stuff to do i might still fire those off with sonnet and that seems plenty viable, but as soon as its an opus tier task i swap to this workflow.
Little annoying as now im kinda trying to manage a .claude/ and an .opencode/ folder but i kinda just have the .opencode/ stuff reference the .claude/ stuff so its a little less bleh.
I've been keeping within my usage because ive been in a funk a bit, but when i was slightly more worried id sorta just juggle whether claude or gpt would handle writing some initial tests as it did seem to kinda be imbalanced otherwise. seems like gpt just spam resets weekly usage throughout the week anyway so its prolly nbd.
I wouldn't be surprised if folks start complaining to California government agencies like the Department of Consumer Affairs, and they take it seriously.
There is a lot of political capital to be earned by appearing to be "tough" on AI companies.
> Claude has become practically unusable for Pro users in the past few days. The Opus 4.7 blew through an entire 5 hour limit in one question and didn’t even finish answering it
Glad I’m not the only one!
I’ve been limited so often this week I’ve setup half a dozen token compression tools in my workflow and had to do a crash course in token optimization.
Of course, it seems to only slightly delay the inevitable and doesn’t really solve the problem.
I have to guess that they're compute limited somewhere or the new models are incredibly overusing tokens, so I guess you need to wait for new data centers to come online?
I think that Anthropic has capacity problems. They went all in on acquiring new customers but now they don't have enough capacity to both serve users and train new models, so they are trying to limit user usages.
It is pure speculation of course, but I don't have any other explanations on the stuff they are pulling in the last 2 months.
Yeah, this is pretty clearly what's going on, but I wish they'd be more transparent about it. Funneling compute to Mythos and Design, while auto-setting effort levels lower and removing user control of extended thinking. I don't think the need to shuffle compute around is unique to Anthropic, though. I suspect it's part of why Sora got killed. And everyone's having uptime issues. Are we reaching the limits of the available compute?
So their minimum workable offer for devs just went from $17 to $100.
Also, I don't see how the Pro subscription is relevant anymore. Nobody pays $17 a month just to chat.
I just unsubscribed. :)
Time to try Chat GPT Codex, which even works with the free subscription (don't expect crazy token allowance, of course).
If true, very strange change when Codex (at both 20 & 100) is a much, much better deal for a model much better at most coding tasks, with way more usage even with the /fast mode enabled. Is losing most non-enterprise customers the right move for them?
Equally, will offering a presumably unprofitably large quota of Codex tokens at $20 to retain non-enterprise customers turn out to be the right move for OpenAI?
Would not be surprised to see OpenAI follow suit.
Or perhaps OpenAI's LLMs are just so more compute efficient that they can actually offer that sustainably...
Feels to me it's a battle between who has the most compute. OpenAI does not seem to be struggling with their x2 usage on the new 100 Plan, which is very close to unlimited usage with the best performing model on the highest reasoning setting. Not mentioning the resets every 1 million customers, or the other generous usage multipliers last months. Meanwhile Anthropic seems to be desperately trying to cut down on inference with their changes to reasoning effort and more lately, so they might be focusing on what they consider to be more valuable customers for their long-term strategy. The 20 plan with Opus had gotten so bad on CC they might've just pulled the plug to stop people from complaining about usage limits. If OpenAI can burn money longer and capture the market from the bottom, I think they'd win in the long run.
> much better deal for a model much better at most coding tasks
The first assertion is also subject to change (and likely, if this works for A\). The second assertion is very much subjective. I haven't found that to be the case, but everyones needs, use cases, and workflows are different, so glad that it's working that way for you.
That's exactly what I fear- that Mythos/Glasswing has made anthropic confident that they can survive by only serving that type of customer. Would be sad to see.
It cracks me up when I hear takes like - 'if you're not using more than $20, the product isn't for you because you're not a real user.' If you use CC as an assistant rather than a replacement for your own thinking, follow SDD, and use the tool thoughtfully, you deliver a lot more and you don't need the 5x or 20x limit. It's a different story if you're vibe coding, but then we're not really talking about AI-assisted work - your three prompts barely count as doing any work. I've been on Pro for 2 years, but if this is how things are going, I'll look for an alternative. Luckily, there's plenty to choose from.
Seems like a pretty bad business move if it's really what they're doing. They should want devs using the product on a cheaper subscription to see the value with profitable limits on usage.
I think the only reason to do this would be that they just can't scale up to service the volume they have and need to cut down significantly on the total number of users. Seems also like a rough business proposition. Most of the pro plan users would probably migrate to a competitor at a similar price point (I know I will).
The only other possibility would be if they are losing too much money on the compute power and just can't offer it at that price anymore. But then upgrading the plan gives you more compute per dollar, so maybe they're just banking on people not actually using all of what they pay for?
I had previously thought that the inference cost of using a trained model was relatively low and that most costs went into training new models, but maybe that is less true with the more powerful newer models.
If it costs a ton more to serve Opus vs serving something like Kimi or Qwen, then I think most people just won't use the more expensive version for most things.
if this is accurate, and not some "oops we made a vibe-coding mistake updating our website" I am going to hit the "cancel subscription" button so hard that my desk will break in half.
I have an unlimited-usage API billing plan through my dayjob, but for obvious reasons they don't allow piggybacking personal usage onto that. so I paid for the $20/mo personal plan as an easy and relatively cheap method of professional development / keeping my skills current. I don't particularly mind paying $20/mo, but I'm absolutely not paying $100/mo.
also, part of the reason I didn't mind paying for the personal subscription is that I liked having consistency between the tools I use for my dayjob and the ones I use for side projects. if that goes away, then I might as well switch away from Claude usage at work as well. I very much doubt Anthropic's revenue predictions for this change are taking things like that into account.
making a change like this without an announcement, just sneaky updates to product pages, is also an absolutely unforgivable thing to do, in terms of me trusting them as a company.
The age of AI seems to forget some lessons from Google (and history in general).
- Rapid changes hurts the trust of your brand and product. In Google case, using a new service product became something you’ll think multiple times as you are more likely to axe it than rivals or specialized equivalents.
- While models currently has no clear winner. Anthropic’s core product is coding. But just as Skype, IE, Netscape their can always be another game changer you cannot count.
- The Pro plan is already limited for true agentinc workflows. The limits now are so bad that a business that relies on it would need bigger plans.
- Anthropic is already in a delicate situation where many devs are frustrated. Dropping or crippling the use even more just means this sector (which I can only assume is a big chunk) would switch to competitors tool that already try to compete.
- Local models, whether as Google sees it “edge” or even further would also take bigger part in the future.
“The limits now are so bad that a business that relies on it would need bigger plans.”
Isn’t this the goal to some extent? They’ll probably have the standard “light” usage plan for weekend warriors or normal folk looking to play around. Companies that mandate usage and provide the subscription for hundreds of employees will have to cough it up, and will have no problem doing so if they want to compete with the others (or so the hype would allude to).
Claude Design was iterating on the plans page and decided to remove clutter and their review bot LGTM’d it as “minor copy change human review not required” and auto-merged it.
All I want is a reasonably priced subscription combining both coding AI and general AI in a single bill for non professional use that allows me to opt out of my data being used for training.
Google limits history to 72 hours if you opt out of training even if you pay them $20 a month which rules them out for me. I guess I'm going to try the $20 chat gpt plan.
At this point I am wondering if I need to accept that were moving to a token based model and get comfortable with opencode and manually switching models.
Maybe they're putting out a weather balloon to test sentiment. That way when they're caught they can just point at the other page to say it was just a mistake.
Either they baited people with code and flexible usage limits until march and this was planned or they realized that they did too good of an product and it costs them too much.
One thing is clear, Anthropics communications and leadership is horrible. You don't launch or remove features like this. How this is communicated and handle is something like mom+pop shop would do.
Same. I hope we're grandfathered in. Otherwise current pro subscribers who signed up with the understanding that they could use it in Claude Code are going to be extremely pissed and go off and sign up for alternatives (or start running local models instead). I mean, I guess they could say too bad, they got your money, but this would destroy their brand among people who are currently their most loyal users.
Same. I'm not a dev but I use CC a few times in a week and it's been a great help.
However, my company paid for my annual subscription, so maybe I'll ask our lawyers for advice - the only reason they paid for this was my access to CC and with my use the next tier wouldn't make sense, AND no one will expect Anthropic to not nerf it too.
Before you have lawyers look at it, wait until you’re actually impacted. Nothing has been removed from existing subscriptions yet and their employees Tweeted that existing subscriptions aren’t impacted.
Thank for the insightful comment and indeed I wouldn't without the evidence form myself and other people on the same subscription (all non-devs so we use it rarely but it feels like a superpower).
You're right. I didn't scroll down. I wonder why they didn't update the top cards that everyone see. They do it for claude Cowork but not claude code? That is not very transparent. How does it make sense? It's not like claude code is too niche to be included, it's in the main app and I know multiple non-techie people who use it.
I assume this has to do with the $20 tier now running out of provisioned tokens so quickly as to be not particularly useful, giving users a bad experience.
The million token context + reduced caching period + new models using more tokens made this a probably unpopular but perhaps unavoidable development.
There's a hard problem here balancing costs and experience. I'm afraid despite the bad experience for people that this is necessary and $20/month was just too big a loss to sustain.
Is there any marginal cost associated with a new subscriber?
I have always heard inference is cheap and the cost was in training, so I assumed any subscriber was making them money, just not enough to cover their insane fixed costs.
Next they’ll slowly reduce how much CC usage you can get out of the $100 Max plan, then introduce a new $300 “Max Plus” plan with “40x” usage.
“You asked, and we listened: Introducing Max Plus, our biggest plan yet, designed for those…” blah blah
I can't believe they are yanking tool access instead of just reducing the token quota or simply pulling Opus 4.7 access. To be fair even that would be poorly received, but at least people would have a choice of working within limits. Claude Code is their real winner, and a great ramp for newcomers coming into AI assisted development. They are playing straight into OpenAIs hands.
Anthropic NEED to get better at communicating with their customers. The most meaningful updates we get on changes come from employees on X. It's unprofessional and unsustainable.
I would assume users who have an existing subscription will be grandfathered in.
It would seem misleading to sell monthly, or even yearly, subscriptions under the guise Claude Code comes with the subscription, for it to only be yanked out underneath you. (Although depending who you ask, Anthropic have already done actions similar to this).
Anthropic clearly doesn't understand that customers see their brand as "Claude", Google's brand as "Gemini" and OpenAI's brand as "ChatGPT." They have so many plans and exclusions that they risk customer confusion. I was surprised when I was pay $200/month for Claude Code, finding it super helpful, and then I had to pay separately to get API access for an experiment. Why are so many parts of "Claude" separate from each other, especially on a $200/month subscription.
Anthropic better get this sorted out with a proper product manager and marketing or they risk customers jumping to easier to understand platforms that are good enough.
That also was really opaque to me RE: API access. I initially thought at $200/month I could get whatever I needed. I eventually set up a OpenAI API with a few bucks to try what I wanted to.
Would it really be that hard for them to just make all of the changes and then do a redeploy rather than doing them incrementally? It's not like they're just editing the raw HTML sitting on the server manually, right? Actually, don't answer that, I'm not sure I even want to know the answer.
Hey, I'm a pro, and I feel genuinely insulted. I could consider going back to Claude Desktop + MCP, but I'm getting tired of this telenovela, and will probably cancel my sub and take my business elsewhere.
Until you work for a company or government agency that is subject to any sort of technology audit. The moment offshore processes running in China comes up you'll have a never ending hole of questions to answer.
Hmm, we just bought my wife an annual subscription at the Pro tier, largely to use Claude Code. Wonder if she'd be grandfathered in or if we'll need to get a refund.
Sooner or later, Anthropic will probably have to take this step if they want to actually make money.
Either way, they're going to lose a huge chunk of their user base because of it. I currently have a private Pro subscription for $20; I can just barely justify that amount for my personal hobby projects. Work-wise, my employer provides all of that for me, so I don't really care - this is just for my personal needs. But at $100, I'm out. The subscription isn't worth it to me just to run a few prompts in Claude Code once a week (yes, I'm still one of those people who is actually capable of, and enjoys, writing code by hand). I do use Claude Chat as well, but the free plan is more than sufficient for my needs (never ran into any limit until now). I simply have zero interest in Cowork and any products other than Claude Code.
In short: if Claude Code is removed from the Pro plan, Claude in general will no longer be interesting for people like me.
I hope they really botch Claude Code to give all the developers a wakeup call about making a hard dependency to a for-profit entity that's essentially in it's startup phase. Like, we wouldn't adopt some newfangled framework just because it was on the front page of Hacker News. Why did we all jump on the new unproven technology here?
Trying to chat with "fin" is like trying to argue with a bowl of congealed oatmeal.
Me:
Why is your website saying Claude Code is not included in Claude Pro subscriptions?
"Fin", the slow, dopey AI agent:
That’s confusing, but it usually comes down to what “Claude Code” means in that context. Claude Pro does include Claude Code (the Code tab in the desktop app, and Claude Code access for individuals).
What Pro does not include is API access via the Claude Console. Those are separate products: Pro is for chat, and the Console is for building apps with the API.
So if the website is saying “not included,” it’s likely referring to Console/API access, not the Claude Code feature itself.
Did that clear up the confusion?
Me:
No. It clearly states on the pricing page that Claude Code is not included.
Fin:
That’s frustrating, <NAME>. The key detail is that Claude Code is included in Claude Pro, but API access via the Claude Console is not.
So if the page you’re looking at says “Claude Code is not included,” it’s almost certainly referring to Console/API access (the developer platform), not the Claude Code feature itself.
If you tell me the exact URL or screenshot of the page that says that, I can confirm which “Code” it’s talking about.
Fin:
On https://claude.com/pricing, Pro does include Claude Code. It’s listed right there under Pro’s “Everything in Free, plus” bullets: “Includes Claude Code.”
So if you’re seeing “not included” somewhere else on that page, it’s likely referring to something different than the Claude Code feature. The one thing Pro does not include is API access via the Claude Console (that’s a separate product).
Also, the Code tab in the desktop app requires a Pro (or Max/Team/Enterprise) subscription, and it may not show for everyone yet since it’s still rolling out.
Does that clear things up for you?
Me:
there is clearly an "x" at this page. escalate to a human immediately.
and then the cycle continues. I'm cancelling my Anthropic subscription, sick of worrying about limits and getting banned and simply awful experience.
At least for me Claude Code is still working on my Pro plan. I don't know if that's because the change simply hasn't propagated all the way through their systems yet (the change is now up on the main Claude pricing page and on their support pages, but not on the Claude Code landing page yet), or if it's because existing plans are grandfathered in, or what.
In general Anthropic seems to be pretty bad at clearly communicating what is going on. I have both Claude Pro for Claude Code and ChatGPT Plus for Codex, and lately I've been reaching for Codex first more and more often... at least for the hobby stuff I'm using Claude/Codex on, they seem pretty much equivalent in terms of practical capability/usefulness.
This is more ethical than what they've been doing, trying to keep those subscribers but limiting them to the point it's become unusable. But it's also kneecaping themselves because they'll miss out on any innovation and hype coming out of the hobbyist community.
I'm curious about their expectations and how they will interpret the results.
On the one hand, the people there are supposedly among the smartest on the planet. On the other hand, they consistently forget that they're dealing with LOYAL humans, and these humans prefer respectful communication beforehand instead of being messed with every other day.
My hope for reasonable behavior is to not handle it this way. Decrease limits and increase prices if you can't handle it and be _honest_ about it.
Are they just looking for a way to rationalize another hostile act? And already have expectations like:
- "minus 10% in pro signups" -> oh, let's drop those coders who won't pay anyway
- "minus X% in pro signups and plus X% in max" -> awesome, PAY UP!
Max is next. He essentially admits to it in one of his tweets/posts. Explicitly citing it as an example of how they misjudged usage relative to pricing.
ANthropic never wanted my money anyway... they don't allow work + personal accounts to have the same phone number. I had to close my personal account otherwise I could not complete onboarding at work.
You should be blaming your employer for forcing you to use a personal device to access company resources. You should have been given a company phone or stipend.
I did complain about it in our chat and my boss later reached out, offering to set up a number on our company's Twilio account but by that time I had already entered my personal number.
UPDATE: it appears this is no longer a problem. Just tried again today, after verifying the number they ask if this is a personal or work account, I selected "Personal" and now I can use Claude both for work and for my own stuff.
It is perhaps better for Anthropic to do a price hike e.g. $25 or $30 for Pro with a clear/honest messaging e.g. "Running costs are high, price hike is unavoidable" than resorting to these tactics.
These shenanigans are earning them no respect. The market is already annoyed on model serving QA issues, and now (recently) Opus limits. They don't want to lose to OpenAI - understandable - but these shortcuts won't earn them anything either.
The last couple of weeks using Claude has been…interesting to say the least.
Additionally I run a constant hacking contest between GPT and Claude. It’s a toy project and it simulates an attack/defense of a small corporate network.
Claude used to win pretty handily. Suddenly it’s started to lose 90% of the time. I thought GPT had gotten better but no, looking at the logs it seems that Claude is slower and more prone to running in circles. This is still the case when switching to Opus 4.7.
I don’t know what that means but it’s undoubtedly worse.
Everyone that is upset about this should take note: you are not a (coding) customer at $20/mo. Their coding customers spend thousands per month (week!) on claude and it's growing faster than they can keep up with (source: I'm one of them, and I know many other like me. We're budgeting 10-20% of engineering salary spend on tokens). It sucks to no longer be able to code on the cheap anymore, but don't fool yourself into thinking you have any leverage here.
My leverage is i’m canceling my plan. Openai gives me codex+chatgpt for my $20. I use my claude code sparingly, but I enjoyed it and it works great when I needed it. Access to it was a reason they got my money a few months ago and it’s been a shit show of reduction of services since.
Note that some companies, like Amazon, purchased and ran the Claude on their own hardware. They didn't change the model parameters during the Claude Opus 4.6 karma.
If Anthropic continues to getting worse, try Amazon Kiro and other companies that run Claude on their own hardware.
It might be expensive and have a worse experience compared to Claude's code, but at least the model itself is the "original flavor."
Anthropic should be increasing the value and services they offer not reducing it.
It’s the Apple model. Yes you pay a ton more. But my 2013 MacBook Pro 15 I got in college lasted 10 years and was still fine even when it was stolen. That’s what you pay for. You pay for a ton of built in apps and functionality and quality.
Arbitrarily removing things is customer and more importantantly good will hostile.
Oh FFS claude code is the only reason I have a pro claude subscription. I don't even use my personal subscription all that much after spending all day with claude/bedrock at work. I will absolutely cancel my pro subscription and continue to use local / Codex if claude code stops working.
I realize this duplicates a lot of sentiment already in this thread but anyone here with pull at Anthropic please understand it will undo a lot of the goodwill that made Claude so successful in the first place.
Local AI is almost impossible right now with the prices of RAM and GPUs and the sizes of decent models. No way spending even an optimistic 10k, but more likely 20k, on a setup that is good for 5-6 months makes any financial sense.
I disagree wholeheartedly. Older models do not perform anywhere near as well as newer models, and certainly not once you throw in agents that can sense check, security check, refa tor and balance, and research queries as they run behind the scenes.
Impossible to find Mac minis in some areas, and if this goes through expect it to get worse.
I settled for the AMD rough equivalent. It’s not perfect but it can still handle most of the work. Now if only extra ram would come down in price… I find I need about 5 GB more than I have
It’s seems like there are a lot fishy smells coming from the timing of the mythos announcement and the reports of issues with casual users. Combine that with the mass rejection of 4.7 it kinda seems like they are burning their ‘non research’ users in order to keep the Mythos users warm.
I could be connecting unrelated dots here, but it sure as hell seems quite coincidental to me.
To help you decide if you should keep your Claude subscription, you can see how much of your code is written by Claude Code with my project (open source, local): https://github.com/gelatinousdevelopment/buildermark
The clusterfuck about the various pages not being all upgraded to reflect the same story...
I thought we now had advanced tools to which we could ask to do things like: "Remove all mention of Claude Code in the Pro, but not in the Pro Max plan".
But apparently the CGI-days called and asked the webmaster to manually edit .html files one by one?
But the current plans are unsustainable and prices will have to be effectively raised sooner or later:
> Engagement per subscriber is way up. We've made small adjustments along the way (weekly caps, tighter limits at peak), but usage has changed a lot and our current plans weren't built for this.
you could try customer support, that chat bot will happily loop you with some more non answers, but try to make you feel good about those non answers :)
This makes sense given Anthropic’s recent downtime and resource constraints.
Opus 4.7 consumes tokens at a faster rate and folks were complaining that the Pro plan included too few credits for real work.
And Anthropic now allows `claude -p` (which invokes Claude code) for 3rd party agents like OpenClaw, which consume far more tokens by running autonomously, 24/7.
The only thing they'd need to do to enjoy the positive PR from the DoD spat is shut up and improve (or at least not worsen) product.
Even the downtime would've been fine (as GitHub shows). Instead they're pissing it all away by letting employees make random announcements on random platforms.
This is exactly what happens with all agentic products in near future. Demand is way higher than supply right now. Also most programmers do not use agents yet so demand will grow even more. Building data centers is not easy and price of HW is skyrocketing.
I see lots of speculation that Anthropic needs to cut usage because they are compute constrained. If that's the case, will they be focusing on reducing compute costs for their models?
From what I can tell Opus 4.7 is more resource-intensive than Opus 4.6 is more resource-intensive than Opus 4.5.
I would hope that we'd be grandfathered in since we signed up with the expectation that we could use it in Claude Code. I could see where maybe that might lead to problems down the line where they do some kind of update and "forget" that people who signed up before a certain date were supposed to be grandfathered in.
I was seriously considering investing in an RTX 6000 rig, and I think this is just enough to tip me over the edge. I know it's not as good as CC, but at least I know it'll be there tomorrow.
I remember when they first added Claude Code to Pro — it was limited to Max initially — and my first thought was that it seemed kind of stupid, because at one fifth of my current limit, I would be hitting walls all the time...
I’ve found that I hit the limit just around the end of the 5-hour window, so it’s definitely been usable for me.
But I’ve mostly been using it for gitops infrastructure in my homelab. I wonder if the token usage is lighter than if I were developing an application.
It was for about the first 6 months after I subscribed, then the rate limits were tightened to the point of uselessness and pushed me to cancel and go for the Codex plan instead.
I use it on Pro and was just thinking today, there is no way $20 covers the cost of it. But I'm long term unemployed and can't afford any higher tier, so if they drop it guess I'll have to find a non-anthropic solution somehow.
Sonnet in the Claude Code harness is hard to replicate out of the box. That vertical integration is not easily replaceable but by no means impossible...
OpenCode and their Go plan will get you close if you're willing to put in the config work.
For when you do need the larger models Fireworks has a pretty generous 'Pass' that comes out to about $7 a week for some of the larger bleeding edge models.
Other than that Codex's $20 plan is still somewhat valuable though they keep reducing usage. Google's $20 plan will get you some Opus usage in Antigravity and a generous amount of Gemini. Not sure how long that will last as they've been tweaking pricing and planning language recently too.
People really have to internalize that these things are expensive to run, and that there isn't enough compute to go around, like actually for real for real, which is likely the source of this. My guess is its a temporary new sign up pause.
Unrelated to the Claudge Code change, I'm fascinated by people on Twitter and Bluesky posting screenshots of the answers they get from AI like it's an original source of information. It's as if some users see the AI as an authority, and derive some kind of social capital from that authority. For example, in the OP's linked Bluesky thread, one person replies with "Fin says it’s included with Pro" and attached a screenshot from "Fin AI Agent" (which I haven't heard of) that claims Claude Code is still available on the Pro tier. Is that valuable? Personally I don't trust what any AI has to say, especially when the subject is currently in flux.
Another example, I recently saw two people over on Twitter posting LLM responses at each other in a bitter argument about Vercel's security breach. They made no attempt to pretend they'd formulated the ripostes themselves, it was just screenshotting one-sided conversations... What's the point? They could've saved themselves the trouble by spawning two LLMs, naming them "John Doe" and "Fred Doe", then telling them to argue and post the name of the winner.
Disclaimer: I don't use Twitter, Bluesky, Mastodon, etc., so maybe it's not that deep.
The "5x" and "20x" no longer make sense for Max. It's supposed to be 5 times the Pro limits. But if only Max 5x has access, then they need to renamed to "Max 1x" and "Max 4x".
How long until the $10 Github Copilot subscription goes away? That was a great deal for my limited personal programming. The only reason I switched from it to Claude was to get coding and general ai in a single bill.
I think Github Copilot is in the process of slowly winding down right now. They've been putting very, very long (multiple day) rate limits on users for various esoteric reasons for weeks now and just yesterday or so paused signups.
And of course you never want the first tier it's included on because you need actual usage, so in reality you need Max 20x - the price of which I can't even see on the pricing page
Tangent, all those people who are coding in extreme niches, like embedded development, is AI helpful? Can it learn enough from stealing FOSS code to help?
I was working in my biggest project in the last 3 years and i dont got mutch money and they go there and cut claude code from pro like wtf so why am i even paying for pro
* change laws to accommodate it - if corporations train data, on every usage they should pay higher taxes so they can't exploit the open data, but public ledger trained model is fine to use open data
* similar tech exists (bitcoin, torrent), needs some modifications
Yes, it's been a way better deal to go for a subscription than pay as you go for me in the past. I had a month where I burnt through ~3.8b tokens which was somewhere in the ballpark of $8k worth of savings.
Now though I don't dare use spend tokens for basic note taking with Sonnet because I'm hitting the limit over a couple million tokens on the 20x plan, so they've really tightened the purse strings since November.
Team plan shows “Claude code” in a main bullet point still. Which would indicate it is part of the team plan regardless if it has premium seats or not.
But it seems this is all in a state of flux.
And there’s the lovely asterisk at the bottom:
> Prices and plans are subject to change at Anthropic's discretion.
I just cancelled my plan, but still have access to Pro and Code apparently until my cycle would have renewed. Hopefully they get a clear signal from this, especially if more of us cancel with the intention to sign back up should they reverse this decision.
I’d be surprised if they’re running at less than 100% capacity after this. It’s just too useful to too many people for whom an $80/month increase is immaterial (I speculate)
This is not so much about capacity, I pay $X and I get a certain amount.
This is much more about removing access to a feature I have access to. Having a chat with Claude Code (removed) seems to me to be likely less usage than Claude Cowork Tasks (more like open claw, cron triggers).
This seems much more like a SaaS Tax. I would not be as unhappy if they kept the Code feature and gated my access to Opus. I mostly use Sonnet anyway
No particular opinion on this change, but generally pricing is a great way to separate dabblers from serious users. There isn’t a great deal of value in dabblers or what they produce, I imagine that training data isn’t worth much relative to the pro users. Similar pricing story with $100 yearly price for Apple developer accounts that people complain a lot about. The reality is if you’re serious about making something, these costs are pretty cheap.
The folks hurt most by this are serious people in developing countries and young people starting out. Occasionally a dabbler turns into a serious user but I imagine that’s far less likely than people wish it were.
The value to companies who make these changes is they don’t have low value users or low value contributions to worry about, which has its own not insignificant overhead. In the age of AI slop everywhere we’re likely to see a lot more attempts to separate the wheat from the chaff.
The dabbler/serious user distinction isn't the only framing here.
Assuming this limitation applies to team seats in the same way, at $20/mo, businesses could afford to have everybody on the plan. Plenty of folks write only a few hours of code per day—or even per week in their job. These are still professionals, not dabblers.
This just lends more fuel to AI skeptics that this entire thing is a massive, unsustainable grift. The explanation only adds confusion and implicitly means that this was not a mistake. What is someone to take away from this?
That $20/month is not profitable? That Anthropic thinks that people are willing to pay a 400% markup without batting an eye? That Anthropic is desperately trying to clean up their burn rate? Why should we trust a company that can screw up basic PR this hard?
Jesus Christ, it’s literally going to become too big of a risk to depend on Anthropic.
As someone who tries to manage usage for a small team they just added Claude Code to the Standard Team seat now they are removing it!?
Not to mention that they will ban your entire organization from a bot deciding you violated their TOS with no communication and no way to contact anyone to understand what happened.
If this is real we are switching to OpenAI or Gemini it is not worth all this non sense
Huh? I just don't understand why they're doing this. Feels like shooting themselves in the foot, given that Claude's individual subscribers are a large part of who is introducing all their enterprise customers to them. Plus removing access is never good for public perception.
They're trying to find every way to enshittify their partially unprofitable service. When they find a way that sticks, they'll go with it. This has become the preferred way of doing tech business in the US. Create a great thing, give it away for free, hook users in, try to squeeze them. In theory competition should limit this kind of behaviour, but for some reason they big companies all wait on another to start enshittification in unison. How this is legal still puzzles me but evidently that's how it goes.
Here's my hot take: Anthropic et al. are trying to make developing a subscription-only job, and they've done that by illegally pirating pretty much the whole Internet. If they were to go out of business tomorrow and serving models was to become a commoditized service like storage we'd be all better off. Sure, we would have less research on frontier models, but we don't need AGI, we need good local models, RAM and good open source / weight AI tools.
And this is all exclusively about Anthropic. It's insane. On any other tech, there would be a consensus to wait until it's stable, but not AI - we go full throttle when it's AI.
Genuinely curious how people who have implemented this in serious companies are answering these questions, because my answer is to keep it the fuck out.
Saw this coming eventually. $20/month for autonomous agents running 24/7 was clearly not sustainable at API pricing. The part that's surprising is there's still no official announcement - just a quiet page edit.
The $20/mo plan never supported 24/7 autonomous agents. With Opus 4.5 and 4.6 I would hit resource limits after a reasonable amount of work, which corresponded to a variable amount of wall clock time.
This makes me think either they’re severely resource constrained and need to focus on “high value” customers, they’re bleeding money on inference, or their sales and marketing team is incompetent.
Regardless, this feels like a pretty big rug pull. Especially without a phase-out period and a real announcement. As someone using Claude Code on a personal hobby project to get a better feel for its capabilities, I’m not sure what to do now. I can’t justify the $100+/mo plans for a hobby project.
My choices are then:
- Code this project by hand, which would be fun but defeats the point of this being my agentic coding project.
- Find another model and use Codex or OpenCode or whatever.
- Put the project on a shelf till this shakes out.
This was never the case though. There's a per week and per 5 hour quota. If you exhaust either you have to wait for the reset. What they're doing makes no sense.
And yet they're very aware that Hacker News, etc exists and so the awareness and backlash would be instant. It's like they want to get a lower rating from the community. Maybe that's their solution for the resource issue: make enough people mad so they abandon their subscriptions.
“ The standard answer here is no — Anthropic does not typically refund the unused portion of annual plans , and annual subscribers won’t see prorated refunds, retaining access for the full remaining period instead.
That said, your situation is a bit different — you’re not just canceling, you’re canceling because a feature you paid for was removed. That’s worth contacting Anthropic support directly about. Their support team can check your refund eligibility , and this kind of material change to the plan is exactly the case where a support escalation could go differently than a standard cancellation.
You can reach them through the in-app support messenger at support.claude.com or via the thumbs-down feedback button. I’d recommend explaining specifically that Claude Code was a factor in your annual plan purchase. ”
Just do a chargeback its an easy W. I dont see why you'd want to continue doing business with anthropic after a change to a 1 year contract out of cycle
Anthropic’s “Head of Growth” claims this is a “test”: https://x.com/TheAmolAvasare/status/2046724659039932830
This does not explain the changes to documentation.
They later said: https://twitter.com/TheAmolAvasare/status/204672549859272297...
> When we do land on something, if it affects existing subscribers you'll get plenty of notice before anything changes. Will hear it from us, not a screenshot on X or Reddit.
If you don't want things like this spreading through screenshots of X and Reddit, don't run "tests" like this in the first place!
(Also "if it affects existing subscribers" is a cop-out, I need to know the pricing of Claude Code for NEW subscribers if I'm going to adopt it at a company with a growing team, or recommend it to other people, write tutorials etc.)
That tweet only makes things worse. On top of all their other nonsense recently, it actually convinced me to cancel my subscription.
I can't trust Anthropic to manage their products in a way that supports my workflow.
pretty much none of these big providers are offering the guarantees needed to be taken seriously in workplaces right now. the technology itself isn't offering the deterministic guarantees that should warrant it in the workplace right now. problem is everyone's foot is just on the gas. even if your workplace isnt paying for it, people are just straight up rolling their own personal claude accounts to do work at orgs.
ive been trying to make the case all year that if we're going to let employees do shit with ai, lets try claude. in the past like.. 2-3 weeks all that goodwill has basically evaporated.
local inference needs to take off asap because all of these entities actually suck and i wouldn't trust a single sla with anthropic. they are not acting like a serious company right now, this is a joke.
What are you guys subscribed to if not Claude? Copilot? Or is everyone legit bringing their own license?
Copilot's per-prompt pricing model is overwhelmingly the best value for money right now, although they more than doubled the price of Opus 4.7 compared to 4.6 and completely removed 4.5 and 4.6, which erodes their lead somewhat. Copilot restricts context much more than CC, but I still find it to be plenty capable. I've occasionally managed to give Copilot/Opus 4.6 a prompt that kept it productive for a full work day for a cost of just ~$0.10.
My company has Github Business, which provides all major models. Claude runs out of tokens usually in the middle of the month. I can use it with opencode.
Then one extra Claude Pro. Then also:
Codex business for a test month. gpt-5.4 was excellent.
And since kimi-2.6 I bought their monthly for testing. Works fine. Not as excellent as opus, but usable. Fixes tricky problems by its own.
Not an OpenRouter subscription yet. Those models are either free on opencode anyway, or not good enough. The free MinMax 2.5 on opencode is pretty usable also.
I got a claude api key from work that I use with zed. It works really well and usually ends up costing $25/day if I use it all day.
I did recently experiment with copilot for personal stuff, but will cancel it now that opus is no longer available on the $10/month plan.
I have the basic subscriptions for Copilot, Claude and Codex. About €50 per month.
I enjoy Codex the most
But like Claude I’m not loyal to any of them.
Anthropic is absolutely taken seriously in workplaces, what are you even talking about?
No serious business uses Pro or Max, they are all on Anthropic API billing.
In fact with this move it is plainly obvious that Anthropic is moving compute from prosumers towards enterprise.
I know of a very serious business that deployed Max to all of their developers. API pricing, from what I see, can become more expensive than just hiring another dev.
We're also not seeing much difference in real throughput at an agency. Everyone is getting decent results, output wise but it just doesn't seem to change the outcomes that much. There is also a mixed incentive at an agency, because a reduction in hours spent is a reduction in revenue.
It will be interesting to see how it all plays out, but I suspect if cost continues to increase and output only improves incrementally from here, that the cost will be the final decider rather than the competence.
I could see it being a thing we use only sometimes, for some things, but ultimately remain reliant on developers to get the work through the pipeline.
API usage is on-demand, employees are a constant cost, guess what management loves most.
Management seems disconnected from reality. Real employees accumulate tribal knowledge, have an almost infinite context, and don’t keep disabling unit tests because they don’t pass. They don’t really cost money if it’s information workers that build almost all of the modern service industry. It’s management that we should see as a cost center.
That's true but employees offer more than code output, and you still need people operating the "machine" at this stage.
I am interested in how corporate politics evolves in this new environment. Usually all the way up the chain, managers and directors use head count as a measure of power and influence (and compensation). Who's going to pay a director top level pay when all they're doing is funneling requirements to various agents? That seems like a technical role that isn't particularly aligned with the soft skills management excel with either.
Well yes it is expensive, but companies are paying for that. It is far more expensive than the Max and it does go up to or more in some cases compared to the employee salary.
Larger companies are using Claude through AWS Bedrock and are willing to easily pay $5k+ per engineer per month for it.
The thinking appears to be that a model that can do the work of a developer must be worth a significant share of a developer salary. I think this idea is flawed.
Developer salaries are driven up by scarcity - scarcity of developer skills overall and scarcity of developer skills in specific places like California. If AI models destroy the scarcity then the price worth paying for a coding agent will drop dramatically.
Maybe Anthropic can get away with it for a couple of months. But this will not last.
But if e.g. a developer can do 50% more, shouldn't it be worth it to pay up to 50% of developer salary for the product?
So the % is debatable of course. There's cases where an AI agent can save weeks worth of investigation, there's cases where you are mainly blocked due to processes, and many different circumstances. It's up to every company on their own to decide it. But if they decide it's 50%, why shouldn't they spend 50% of salary on it?
Like imagine a large company with thousands of microservices. You need to build a feature, before you had to setup cross timezone team meetings to figure out who owns what, what is happening in each microservice, how it all connects together. But now you can essentially send an AI Agent to scour and prepare all this material for you, which theoretically in this planning could save hours of back and forth meetings.
If 1 hour / 1 eng costs $200, then a 10 people 1h meeting avoided would save $200 x 10 = $2000 alone.
I don't see it as a replacement for dev, it's more of a multiplier.
I believe what GP is saying is that there is a price calculation today, but then if enough devs become unemployed, their salary will go down, making them more competitive by finops calculations, at which point the Ai prices will have to come down as well. Where equilibrium is, no one knows
I think it's an interest hypothesis but I don't think it works out like that. AI prices aren't priced in relation to the work they do, they're priced in relation to tokens (input/output). As long as it's cheaper to use those tokens than it is to pay a dev, then dev salaries will likely fall. Whenever it becomes cheaper to hire a dev than to use AI, a company will likely just hire a dev. But AI prices won't fall just because dev salaries have.
Yeah, I mean I think there's just too much work and I think devs who are effective with AI won't become unemployed, but their productivity will be multiplied. More will be expected of companies in terms of output, so it will be just more output.
>But if e.g. a developer can do 50% more, shouldn't it be worth it to pay up to 50% of developer salary for the product?
That's the upper bound but it's not the market price.
Accounting software (+ hardware) doesn't cost nearly as much as the accountant hours it saves. Accountant salaries are simply not a relevant yardstick for the price that software vendors can charge for accounting software.
Equally, the market price for code generators will not stay anywhere near the price of developer hours it saves. It will be determined by competition.
Because accounting software is cheaper due to competition. In software eng Claude is currently strongest and there's higher costs involved than normal SaaS. There are many fields in which the tools/machinery cost more than the salaries of people.
>Because accounting software is cheaper due to competition. In software eng Claude is currently strongest and there's higher costs involved than normal SaaS.
Yes, competition not salaries determines the margins that software vendors can charge. That's exactly what I'm saying.
My expectation is that competition between coding agents will stay strong and costs for the current level of software engineering performance will fall.
>There are many fields in which the tools/machinery cost more than the salaries of people.
For example?
Agriculture, oil drilling, trucking, etc...
The machines in those sectors do not cost nearly as much as doing the same work manually. Not even close.
Of course not, but they cost more than the person using them, it's multiplying the productivity of that person, so if AI multiplied enough as well it would make sense.
That's beside the point.
The question was whether the salaries that would have been paid for the working hours replaced by the machine are a realistic yardstick for the market price of the machine.
That's clearly not the case in any industry.
In my calculation a good opus developer can do 10x more, not just 50%.
Got all my tickets from the last two years fixed on a few days. And implemented all the ideas which came to my head.
But the API is incredibly expensive. I calculated that I would have spent 3000 EUR the last month, a lot more than the 100 I pay now.
Nothing for large companies though.
I am pretty sure that a hole in the pocket in the order of 50 000 000 USD/month (assuming around 20 000 people using AI in not the smartest or most optimized way possible, therefore burning A LOT of tokens) will be noticeable by even the largest companies.
It is noticeable and even promoted, large companies do pay such sums for the API, like $5k+ per person per month. Not every eng is using AI that much already, but companies are clearly willing to pay those sums.
Per employee?
This not nothing.
My startup just rolled out MAX plans for our engineers. we literally jumped into AI last week. Been in operation for 6 years and are profitable.
I work for a "very serious" company with many billions of dollars of revenue. All our SWEs have max subscriptions.
I just cancelled before seeing this news. i was already pissed about constantly hitting limits on the 20 a month plan and looking for alternatives and this seals the deal. Bye bye!
Yea, I've been fine so far, but something happened with Opus 4.6 and especially 4.7. I was able to do some actual work with a Pro plan before. Now it's just pure anxiety of hitting the limits.
With Sonnet it's a bit better, but I can get the same performance with GPT-5.4.
Now I'm pretty much paying the 20€ for Claude Pro so it can plan/review stuff and then I use pi.dev + GPT-5.4 for the actual work.
I just paid for Pro for the first time 24 hours ago. Its been great, but the limits are crazy. It's nice not dealing with ChatGPTs sycophantic gaslighting, and not having random bugs.
That said, I seem to be caught in that 2% test if I open in a private tab. What nonsense. I wouldn't be paying for Claude if it wasn't for its quality abilities, which necessarily includes Claude Code.
I can easily hit the weekly limit on Claude even on the $200 plan. I have yet to ever hit a rate limit on Codex $100. And the results are almost as good. And don't get me started on Anthropic's extra usage scam.
With Opus 4.6 on the $20 plan the limits were bad, but at least you could do a short session.
I find that with Opus 4.7 I can do two messages. Once I had a short session with 4-5 messages and it consumed $10 in extra usage.
This relegated Claude to a backup option in addition to Codex, which has the better desktop app anyway, and much better usage limits.
I’m considering to even cancel Claude entirely.
How do you hit that limit? I’m never close to it.
Not the op, but it’s fairly easy to hit if you automate a kanban and have some stuff you want to get done. All those little “wouldn’t it be great if” tasks that show up after doing a big task become very doable, it just soaks your tokens.
A/B tests only work if the subjects don't realize they are in a A/B test.
Perhaps vibe coding the A/B testing engine isn't the best idea.
Solution: don't A/B test your users.
A/B testing people without their informed consent is immoral, unethical, and should be illegal.
To play devil's advocate, without A/B testing a lot of decisions would be made with insufficient relevant data, and lead to subpar results that affect the many negatively form the road.
counter-point : the companies that are most famous for A/B testing routinely are also the ones with the most notoriously non-existent customer service departments globally, facebook/google/amazon/ebay. Groups that harbor dissatisfied customers by essentially being 'the only show in town.'.
so, what i'm saying is : I think a lot of companies align themselves with the cash first and then measure whether or not the negative image/user impact is manageable .
(in fact I know they operate this way.)
A lot of decisions made with A/B testing are also made with insufficient relevant data, but it's less obvious since it's easy to think the A/B results cover everything.
So you're saying software should never change or you're happy with A testing, but not A/B testing
> So you're saying software should never change
Generally, yes. Make your software better first before releasing it and you won't need to make changes to it.
Want a new feature that you didn't have before? That's a new software product.
> or you're happy with A testing, but not A/B testing
I'm happy with testing when the user has explicitly opted-in for it.
It is necessary to have a control group, just as in trials for new drugs.
Depends entirely on the stakes and whether personal data is involved
> Depends entirely on the stakes and whether personal data is involved
Sure. Let me just A/B test whether or not you'll respond positively or negatively to having your news delivered via push notification or delayed by 10 minutes.
I'm sure you would appreciate being tested on without your consent, just so that I can make an extra quick buck at your expense. Nothing amoral or unethical about it.
What do you think about slow rollouts for new features? Like, we think this new push notification system will be loved but let’s ship to only 1% of users in case there’s a horrible unforeseen consequence like occasional 10min delays? Dashboard goes upside down -> revert then work through logs to figure out what the hell went wrong.
What do you think of things you purchased changing over time into something you didn't purchase?
That's literally any software subscription ever.
So you're perfectly okay with repeatedly paying for a shit product, getting shat on by the company in the form of being tested for feedback, and "maybe" getting a better product in the future. Mind you, that "better" isn't necessarily better for you but more explicitly better for the company you're paying.
Sounds like someone who doesn't care about being a sheep. Or maybe someone whose salary depends on having sheep.
I am not getting shat on by them working to improve the product. Have you got a screw loose?
I think you are making far too wide-sweeping statements. I think most people here probably agree that if Anthropic drops Claude Code from the Pro plan after people have paid with the understanding that it is part of the package, that would be wrong, and they deserve to lose business over it. However, there are plenty of situations where A/B testing is entirely benign, and I would not have any problem with a company doing that testing without getting consent first. Every form of A/B testing is not done just for the gain of the company doing the testing.
Well, duh, that's precisely what's wrong with subscriptions.
That the product might improve?
In my experience, improvements are so rare compared to regressions, they might as well not exist.
How does your example warrant such moral righteousness? The stakes of a late news notification seem strikingly low.
Agreed and I can't wait until they regulate this stuff out of existence. It's absolutely hostile software technique that is deeply anti-human.
> Will hear it from us, not a screenshot on X or Reddit.
Has this ever been true? You will almost always see some anecdotal screenshot a long time before any company would rat on themselves.
Yes the random screenshots include a lot of false positives. But official comms have a lot of their own problems given how companies behave nowadays.
> I need to know the pricing of Claude Code for NEW subscribers if I'm going to adopt it at a company with a growing team.
I agree, but can you really use Claude Code on the Pro plan as a full time developer, or professional 'knowledge worker' without hitting the usage limits fairly early in the day anyway?
It depends on the kind of work you do.
I'm in the academia, and Claude's performance in my field could be described as a very fast junior grad student. When I use Claude Code, I typically spend a few hours figuring out what needs to be done exactly, and describing it in sufficient detail. Then Claude does it in 30 minutes, while an actual student would need days. And then I spend anything from minutes to days evaluating the results, depending on if it needs to be tested with real data and how much weirdness those tests uncover.
But I also have other work to do beyond guiding the automated grad student. Which means my Claude Code usage rarely exceeds 1–2 hours/week.
I use Pro professionally and didn't hit limits most of the time. I believe I used up 5hr quota once or twice. We switched to Team sub and I'm on Standard(which is Pro x1.25 I believe). I don't vibecode entire applications, I ask it to make boilerplate, smaller, well scoped features or fix some errors. I don't let it go off with a prompt "make another netflix clone" cause I just don't see any real value in that
Just the Pro Plan Claude Code on its own? Maybe you could last a full day on just using Sonnet. Maybe one Opus dab in the morning to plan your Haiku/Sonnet day?
I have Pro Claude, Plus GPT and Pro Gemini. When one runs out I switch to another project on the next LLM. If I really need a task finished I'll restart it on another LLM, but I'm loathe to do that as it eats tokens just getting back up to speed.
I think it's more about how they approach their users in general that is the problem here.
It’s pretty reasonable to say “demand is way up, quality is up, supply is constrained, and so price needs to rise”.
It seems weird to segment this way though. Surely it’s better to just give Sonnet to your bottom tier, rather than cut out the entire Claide Code product entirely?
Give folks a taste rather than lock the whole product behind a $100/mo plan.
But if Sonnet is bad it would give bad impression of the product, no? And it also takes compute, so you give a bad hallucinating impression of your product while still losing compute.
It’s not bad though, it’s crazy good in comparison to any model older than 1y old. If you don’t have access to any vibe coding at all it’s gonna be life changing.
But I think you are right, as long as Codex and Gemini are cheap alternatives then vs. 1yo models isn’t the correct comp.
Then it’s probably better to just resegment the whole Claude Code product as an enterprise only tier. (That also has the advantage of kicking out all the Claw subscribers that screw over the token limit economics for normal $20/mo users.)
I mean, this is why they do A/B testing. This way of testing stuff is not new at all, people who act genuinely surprised need to do a reality check. Companies want to maximize profit. They do this by testing what creates the biggest profit. A/B Testing is one of the ways to do this, and it has been used for decades in precisely this way.
Haha, right, just like the recent uncommunicated changes to limits, cache, etc.
A screenshot is, however, apparently good enough for _new_ subscribers.
> I need to know the pricing of Claude Code for NEW subscribers if I'm going to adopt it at a company with a growing team
Subscriptions aren’t for company use.
Just like they gave plenty of notice regarding OpenClaw?
Maybe a silly bet where the head of sales had 1-2 glasses of wine too much... "I bet they will still pay us 20 bucks/mo without CC! Don't believe me? I'm going to prove it!"
[dead]
> So we're looking at different options to keep delivering a great experience for users.
his title should be changed to Head of Corporate Bullshitting
>"his title should be changed to Head of Corporate Bullshitting"
They're hitting the physical limits of energy production and chip supply for inference capacity. There's literally nothing that can be done but reduce usage to spread it around for now.
there's nothing stopping them from saying that, which is my point which you missed
why do they need to say it? It's obvious
No, it's not, except for a small percentage of the userbase.
Hopefully the negative responses in that thread + the conversation here on HN might help them realize that totally removing Code access for Pro users isn't a good look.
And with no free trial period on top of that, nobody is going to want to pay $100+ just to check it out. I can't imagine the conversion rate of that test being positive.
A few enterprise customers I know are upgrading to the higher plan now that their limits have been nuked.
I imagine Anthropic is trying to see how many users they can push to higher tiers with these new squeezes.
I hate to say it but I imagine it will work.
It’s going to suck for me, because I had gotten used to ridiculously cheap tokens, but I guess the era of subsidized tokens is over.
I would guess that even now, they’re still subsidized. Just judging by how desperate these companies are to get ahead of each other
good reminder to me to stay in practice with manual coding for my side projects! claude is super convenient for them for now but if it goes it goes, I definitely don't want to get dependent on it. maybe the local models will improve in a year too.
Most real businesses are on API billing, not Max.
None of the companies I deal with (~40) are on the API. Some where, but that experiment lasted a month.
Until they go public, we are all just guessing.
CC has such egregious API subsidies that it’s hard to not to leverage it unless the license tells an enterprise otherwise. Love the subsidized pricing while it lasts.
> CC has such egregious API subsidies that it’s hard to not to leverage it unless the license tells an enterprise otherwise.
It's hard to tell, honestly - about half the HN population will tell you that all the token providers are running inference at a profit when using the API and only the subscriptions are subsidised, while the other half will tell you that everything, including both the API and the subscriptions, are subsidised (i.e. running at a loss).
Some days I really wish the timing had worked out for me to build a serious private inference server before hardware prices spiked.
What API subsidy? What are you talking about?
My company currently uses the Anthropic Enterprise subscription plan, but we’ve been informed that’s going away in 2027 in favor of API billing. If businesses are using subscriptions, I don’t think they will for long.
You mean the Team Plan? The Enterprise Plan has no usage contingent included.
If your definition of "real businesses" is "Fortune 500, US based tech company with more money than sense or just happy to bleed VC money", sure, 99.999% of businesses are not real businesses.
You may also have a very narrow view of how the world actually works, left as an exercise to the reader to figure out which one it is
Fortune 500 and VC money are disjoint sets.
You have said this in a few places throughout the thread. At this point, citation is needed.
I work for a real business and switched from API billing to max+overflow. It saves money. It’s crazy not to. What are you talking about?
[dead]
I think they're at that stage where people know they want it so lack of a trial isn't a deal breaker per se.
I don’t get the surprise or discontent. People hooking themselves up to a paid SaaS that only two vendors can offer (Anthropic and OpenAI), no competition or regulation to speak of… of course they’ll do whatever they want with their plans.
Hope you can still resume working on your projects without AI.
> Hope you can still resume working on your projects without AI.
How often does technology really move backwards? I can already run a decent local model on a spec’d out MacBook and for better or worse electronics/computers moves one direction (with minor supply chain hiccups along the way notwithstanding)
Losing trust on them not rug pulling users
If anyone was paying any attention to how corporations run, I don't see how they could have believed this would go any differently. Seriously.
If one doesn't want rug-pulls, one signals their policy makers to create regulation to prevent it. Otherwise it's just... uncapped capitalism or what's the name
> on ~2% of new prosumer signups.
I, and everyone else I have asked, see this new updated sales UI; sounds like more than 2%.
Either they vibe coded a test that was extremely broken.
Or they vibe wrote some bullshit to try and back pedal.
Yeah I flat out don't believe the 2% thing. It's possible that I was the 1 out of 50 who checked the page and saw that Claude code was removed... but it really seems like everyone I shared it with saw the same thing which is incredibly unlikely. Also I am an existing subscriber and checked the price page while logged in, so I shouldn't be counted in "2% of new subscribers" at all...
He goes way beyond saying it's a test, he's legitimising the change in the follow-up rationale
I am confused, how is this a test? So some users get Claude Code while others don’t, when they are both paying 20 dollars … ? Wat
It's a test on sign ups, not on users, so "will they sign up without X feature for the same price" yes
Just checked. I continue to have Claude Code with my Pro plan
This is concerning though. If I lose my current usage allotment at this price point I will likely switch to codex
The cheapest plan for both Claude and Codex is the sweet spot IMO.
It also forces you to keep your workflow mostly harness-independent because Claude supports next to no standards and Codex does some.
That works until openai does the same thing. Pretty clear as an industry they want to establish a new price floor for non-trivial coding use.
Yep, and the price point theyre looking at is 95% of an engineer.
Once they get people hooked, deskilled, and paying, the money ratchet only tightens.
And the companies KNOW that theyre replacing engineers, or trying to. So each engineer replaced is X salary a year they now have available, so make it back in SaaS LLM tokens.
Thank god for the Chinese labs. Keeping us (relatively) honest.
That's what Claude is testing I guess (people often don't do what they say they do when it comes to pricing)
They confirmed that it does not affect existing users
Presumably for new subs.
This test would be a good way to lose existing subscribers perhaps.
Presumably they want to lose existing subscribers because it’s way too expensive to keep them at $20.
Is it? I’m curious because I thought they were raising prices to pay for exorbitant training costs, not because subscribers are expensive on a unit basis.
I thought inference was cheap so there was little marginal cost of a new subscriber.
This was more or less true until everyone and their dog started running agents in a 24/7 busyloop as a bit.
You’re not doing that on a $20/mo plan.
How can you run the A/B test with mismatched documentation?
It is honestly truly fucking incredible how corps still find new, innovative ways to enshittify. Regular enshittification won't cut it, they have to exercise their artistic creativity. Who the fuck comes up with the idea that what services you get with your subscription are random? It's mind-boggling that some percentage of people visiting the website will be presented with an inferior version of the same subscription for the same price. I'm not even mad (despite my colorful wording), I don't use Claude, just impressed with the bold new territory being explored here.
Claude subscription became non deterministic too
I find the whole thing a bit sad but you made me smile. Thank you.
I think of enshittification as "we're making plenty of money but let's make more." In other words greed.
Based on how much money Zitron has reported that these companies are losing on every subscription, this feels more like they're just trying to survive. In other words "ohshittification."
> In other words "ohshittification."
Brilliant coinage, if it’s yours, congrats!
My take: it is not enshittification to raise the price for a product whose demand outstrips its supply. That is basic economics. There are alternatives, it’s not a monopoly. If you think it’s the best product, then pay more for it.
Personally I would be perfectly content if the price of Max went up a bit and Pro no longer worked for CC if it meant that Max was faster and more stable.
Zitron is completely full of shit too though. I imagine they’re compute limited and so they’re moving towards price discrimination.
> It is honestly truly fucking incredible how corps still find new, innovative ways to enshittify. Regular enshittification won't cut it, they have to exercise their artistic creativity.
I had a bit of an epiphany the other day thinking about these VC companies offering products to the public at unsustainable prices. It's classic anticompetitive behavior.
You imagine anticompetitive behavior to come from a monopoly because they can afford to burn money to drive competition out before they bring prices back to profitable but the whole VC burn is the same thing. People talk about it a lot without really saying it explicitly when they talk about moats. The only moat Anthropic and OpenAI have is money and they utilize it by offering products below cost.
The two companies are just trying to outlast the other one until they are the only one left.
So it's not really enshitification as much as you were previously getting the deal of a lifetime.
In physical markets we call this kinda thing dumping and it's often regulated. Maybe offering SaaS or compute at below profitable rates should be investigatable too, to avoid killing competitors too easily?
Dumping is typically used in the context of international trade.
There are some predatory pricing laws, but they're much more narrow than most people believe. There is no law requiring things to be sold for more than it costs to produce.
I think it's funny that these topics make people angry enough to demand that we make laws to force companies to raise prices. We'll stick it to these companies by forcing them to charge us more! That will show them!
Such laws would be very bad for startups and newcomers because they'd be forced to price their new product higher than established competitors who have economies of scale. It would be a nice handout to the big companies.
> Dumping is typically used in the context of international trade.
This is dumping and it is international trade. Maybe you don't realize it because you're American and have internalized it as business as usual.
The whole Silicon Valley VC industry and the majority of the net worth of SWEs on HN is based on dumping. "Burning VC cash" is transparently dumping, and it's squarely what the US big tech dominance is founded on. Amazon, Uber, Youtube, now LLMs. The huge majority of "success stories" of the last 15 years are based on dumping their product far below cost price, running at a loss for years until they dominate the market, and then jacking up prices/enshittifying/selling user data.
I didn't think Amazon engaged in dumping.
It took about 9 years of losses before they got profitable, which is the exact mechanism I talked about.
Dumping and losing money overall are different. Selling $100 bill for $10 is dumping.
2 day shipping is dumping i guess
Yeah, that's where the realization led me too.
These companies probably need to be forced to at least try to price their products at a level that would be sustainable long term.
This happens naturally because no company can run at a loss forever.
I think it's funny that we're getting subsidized and discounted services and this makes some people so angry that the comment section is demanding laws that would force companies to charge us more.
Of course but they can run at that rate long enough to make it impossible for another company to compete and go bankrupt. Which is the problem.
Like I said, this is the iconic strategy of monopolies. They take their pool of money, move into a new market, lower prices below cost until local competition withers away then buy their assets at bankruptcy prices and raise prices to whatever they want. Those temporary discounts are designed to kill competition, innovation and choice.
VC subsidized prices are exactly as bad for the market as that. It's unclear to me if the intent is the same but intent doesn't really matter.
I'd also like to state that I'm not angry about it.
Wait what, so they're testing giving new users misleading information about included services in each tier as an upsell tactic?
It could be an A/B test to see whether people without an existing subscription care about Claude Code (CC) at all. If they sign up then CC is disabled (or not as it is not really an issue to offer more). Capturing that info would definitely be useful to a growth team.
They need to give the non-code service a different name.
And a slightly lower price.
If it succeeds they can adjust pricing later.
Otherwise they are messing with their new and old customers heads, regarding a service with a name that ought to be reliably interpretable. And seriously messing with their own credibility. Wrong kind of A/B test.
This is incompetence which i would normally discount. But Anthropic seems to be falling all over themselves to irritate customers.
No, they're testing removing it from the Pro tier for new subscribers.
No I think the test is that some new sign ups won't get Claude code in that tier if they pick it and they're seeing if users will still pay for it without it?
I think the test is that new sign-ups won't have it and will the loss of five new Pro subscriptions be offset by more than one new Max subscribers.
Plenty of Pro subscribers never touch claude-code.
Although the ones that never touch claude code are a free $20 a month, the ones that do are potentially a seventy to eighty dollar twenty dollars a month . it’s not instantly obvious which customers you prefer (revenue vs cash negative growth- on second thought obviously they prefer the second)
They've preferred the second so far, but they might have a fair reason to see if they can keep growing with the first one instead or cut down on some loss leading, right?
That's how i read it too - they want to test if people will still pay for pro plan if it doesn't include Claude Code. At the same time they are also saying that if you subscribe having been told it does include Claude Code, they may still change their mind later and take it away!
A/B test to see how people react to changing the offerings
5000 angry users who leave? maybe keep it. 500000 users leave? pretend it was just documentation mis-prints or an oversight, etc.
Somehow a ton of people are caught in the variant.
Random data point: Guest passes apparently still include Claude Code in their Pro trial. If they are running a test this is a really sloppy way to do it.
Fk around and now they will find out
This reeks of the start of enshittification. Very doubtful it was a "test"
Soooo Sam Altman replied “ok boomer” to that message. Wtf?
[dead]
Apparently it's just an A/B test. Legit LMAO moment, speedrunning reputation destruction to your entire userbase just to test a question whose answer you can probably already guess.
---
> For clarity, we're running a small test on ~2% of new prosumer signups. Existing Pro and Max subscribers aren't affected.
> When we launched Max a year ago, it didn't include Claude Code, Cowork didn't exist, and agents that run for hours weren't a thing. Max was designed for heavy chat usage, that's it.
> Since then, we bundled Claude Code into Max and it took off after Opus 4. Cowork landed. Long-running async agents are now everyday workflows. The way people actually use a Claude subscription has changed fundamentally.
> Engagement per subscriber is way up. We've made small adjustments along the way (weekly caps, tighter limits at peak), but usage has changed a lot and our current plans weren't built for this.
> So we're looking at different options to keep delivering a great experience for users. We don't know exactly what those look like yet - that's what we're testing and getting feedback on right now.
> When we do land on something, if it affects existing subscribers you'll get plenty of notice before anything changes. Will hear it from us, not a screenshot on X or Reddit.
https://x.com/TheAmolAvasare/status/2046724659039932830
---
Personally I love how they have increased everyone's quotas to counteract the Opus 4.7 tokenizer change a few days ago, but are immediately regretting it and trying to cut off subscription users.
If the subscriptions are unprofitable, then just communicate honestly, raise the price or lower limits for new subscribers transparently, and grandfather in existing users. That's what GLM coding plan is doing and it works fine for them. Don't ruin your reputation with opaque messaging and hidden changes. Lol
> If the subscriptions are unprofitable, then just communicate honestly, raise the price or lower limits for new subscribers transparently, and grandfather in existing users.
This. Why do so many companies fail to get this? Anthropic's user base, in particular, is intelligent enough to understand their constraints.
It's possible they don't know the actual costs. The one-time costs like hardware and training foundation models is huge. There's the ongoing costs like the PhD dudes they hire. There's the data they buy and decide not to use. There's the various offers under one payment plan - limited artifact hosting, cowork, image upload, Claude Code, Code tab on Claude. What about feature parity between the site and app? Who's working on all these?
tldr it seems really complex and by the time they've counted it they probably hired 40 new people for an unannounced feature.
I mean, if you look at the Claude subreddits, the general consensus is that Anthropic and OpenAI are money hungry corporate devils that are here to enslave all of us, and everyone should get unlimited everything on Max x20.
I think you may be overestimating the willingness of people to understand Anthropic’s concerns.
Reddit is a very small subsection of the Internet consisting mainly of basement-dweller radical leftist they/thems. We should not be taking them as a representative sample of the average Anthropic user.
Ironic that you are talking about unrepresentative samples while characterizing Reddit users that way. Reddit is a huge subsection of the internet, over a billion monthly active users. It covers basically all strata of western society you can think of.
They seem to count unregistered users in the D/W/MAU, but even so the producer / consumer of content ratio across social media still stands. What you read is a subsection of all visitors to Reddit, which follows the power law.
Are you counting the fakes and the bots?
Moreover, as sibling pointed out, content creator != visitor.
If Reddit was a representative sample of reality, all governments in the West would have been ran by gay race Communists.
I have to ask - are you a bot? You seem to be making some very strong and misinformed statements here. You can fixate on the number if you like, but calling Reddit homogeneously radical left or 'very small', is pretty stupid.
Thinking that Reddit has any resemblance to reality is what is stupid.
big fan of A/B tests that dehumanize the consumer into some kind of money making lab rat funnel whose only purpose is to be experimented on how you can extract more money out of it
peak siliconbromaxxing
While I feel the same way, this is nothing new at all. Basically every company does this and it's a totally normal way to test new profit models. Has been done for decades. People acting surprised here really need to get on with reality.
Stage 2 of enshittification. A necessary prerequisite to full enshittification is lock-in, so be on the lookout for it!
But they have red ethical redlines bs
MS paused Copilot subscriptions because they don't have enough capacity. Anthropic is trying to confuse new users and literally don't want 20$/mo because they don't have enough capacity. Seems like there's a trend here. A lot of people in scaling threads were saying that capacity projections and DC buildouts were "fantasy" a few years ago. Not so much anymore...
What does it mean that they’re running a test? If you’re one of the unlucky 2% you need to pay more?
Usually A/B testing is just on the surface, and when you actually subscribe you get the "better" terms of the possible options.
Like, they're just advertising different terms to test how many people would still click on it and very likely start the subscription process, but after they click they go back to the usual terms. Changing the whole payment flow, account models and permissions in their backend just for a quick test is usually too much work.
But yes, basically, if you're B and not A, and B has objectively worse terms than A, then you're just unlucky. But this is the essence of A/B tests. They are done by basically every company everytime, because it's the most straightforward and simple way to test new terms or designs.
I don't think anyone is questioning the idea of A/B testing in general, but we are confused how it works in practice when you are advertising a service. Can you give me an example of this happening elsewhere? Or are you saying it's so normalized that there would be no record or news articles about it, and this only is newsworthy because "AI"?
I have never heard of a company intentionally advertising less features to only some of their customers to see if they'll still bite.
And we still have no actual confirmation that Claude is giving people who fall into that 2% of testers the full service, as opposed to the advertised one.
If it’s just a “test,” why did they update the documentation?
Actual lies - the documentation was changed.
I don't think I've ever been on such a rollercoaster with a company's reputation in the developer space. I started in January on the $20 plan, essentially my first agentic AI programming. I quickly started hitting limits developing several apps at the same time. I went up to the $200 plan after seeing the value.
After seeing my own issues with 4.6 and the mega-post on Github about declining metrics in a decent dataset of claude chats by Stella Laurenzo at AMD (https://github.com/anthropics/claude-code/issues/42796), I downgraded to the $100 plan. Hallucinations. Laziness. Lack of thinking. The responses on those mega-threads from Anthropic rubbed me the wrong way in a "you're holding it wrong" kinda way.
In the past week, I downgraded back to the $20 plan because the Codex $20 plan on 5.4 was working so well for me.
Then throw in other oddball events like the source code leak, and the super positive Anthropic events like their interactions with the current administration. It's a wild ride.
I can't understand removing Claude Code from $20. I'm interested to see whether this is confirmed or not.
I'm a career engineer and I went from being one of their most outspoken proponents (at least within my circle) and now.... I'm not.
Same loved them, told my team about them, got them to switch off of cursor, now I'm telling them to swap to Codex.
Anthropic really pissed me off with their harness crap. They're well within their rights but their communication over it was enough to get me to swap. I don't need extra hurdles when there's a perfectly valid alternative right there. They don't have the advantage they think they do.
I think we are inevitably heading to using the cheap Chinese models like Kimi, GLM, and Minimax for the bulk of engineering tasks. Within 3-6 months they will be at Opus 4.6 level.
This was literally my task today, to try out Qwen 9B locally on my, albeit a bit memory-constrained at 18GB, macbook with pi or opencode. Before reading this update.
Minimax coding plan is $10 a month for roughly 3x the $20 Claude Pro CLI usage allowed. That would be good place to start. 200k context though.
MiniMax has its own issues. Server overloads, API errors, and failure to adhere to even the system prompt. It can happily work for hours and get no job done.
Just like me :)
Please report back, would be very interested in your findings.
I ran OpenCode + GLM-5.1 for three weeks during my vacation. It’s okay. It thinks a lot more to get to a similar result as Claude. So it’s slower. It’s congested during peak hours. It has quirks as the context gets close to full.
But if you’re stuck with no better model, it’s better than local models and no models.
I have to say, OpenCode’s OpenUI has taught me what modern TUIs can be like. Claude’s TUI feels more like it’s been grown than designed. I’m playing around with TUI widgets trying to recreate and improve that experience
> I have to say, OpenCode’s OpenUI has taught me what modern TUIs can be like. Claude’s TUI feels more like it’s been grown than designed.
Claude's TUI is not a TUI. It's the most WTF thing ever: the TUI is actually a GUI. A headless browser shipped the TUI that, in real-time, renders the entire screen, scrolls to the bottom, and converts that to text mode. There are several serious issues and I'll mention two that do utterly piss me off...
1. Insane "jumping" around where the text "scrolls back" then scrolls back down to your prompt: at this point, seen the crazy hack that TUI is, if you tell me the text jumping around in the TUI is because they're simulating mouse clicks on the scrollbar I would't be surprised. If I'm not mistaken we've seen people "fixing" this by patching other programs (tmux ?).
2. What you see in the TUI is not the output of the model. That is, to me, the most insane of it all. They're literally changing characters between their headlessly rendered GUI and the TUI.
> Claude’s TUI feels more like it’s been grown than designed.
"grown" or "hacked" are way too nice words for the monstrosity that Claude's TUI is.
Codex is described as a: "Lightweight coding agent that runs in your terminal". It's 95%+ Rust code. I wonder if the "lightweight" is a stab at the monstrosity that Claude's TUI is.
To be clear, was OpenCode a better in your opinion compared to ClaudeCode?
Better UI, worse model (GLM), probably slightly worse agentic runtime.
In spite of how glitchy Claude feels, it makes decisions fast.
For what it's worth: here's my experience in the first 10 minutes of using Qwen locally to write some code. https://github.com/robertkarl/local-qwen-first-10-minutes it includes some token generation numbers and steps to repro.
how was it? I'm doing this today
I will report back... but I have to recommend this comment on a post about Qwen 3.6 https://news.ycombinator.com/item?id=47843466 by daemonologist
it goes into detail about llama-server args; quants to try; and layer/kv cache splits. I plan to try the techniques there.
Kimi K3 in July-September is the big one.
Kimi 2.6 works roughly like Opus 4.6, when it used to work. Depending on the task, a bit better or a bit worse. And it's MUCH cheaper.
From this morning: I had a single go file with like 100 loc, I asked it to add debug prints, it thought for 5+ minutes, generating ~1m output token and did not actually update my file.
Which harness? Did you use OpenRouter?
Anthropic will kick and scream as those are often distilled from their latest models and is cutting into their margin. Though it is not like their hands are clean neither, it is just a different type of stealing, an approved one :-)
How challenging are these to setup locally and have them running?
Getting them running is easy (check out LMstudio or ask one for some recommendations). The real question is whether you have the hardware to make them run fast enough to be useful.
The min req is probably crazy I assume but I'll take a peek :)
This is possibly a hot take but recently I've been having about as much luck with Composer 2 in Cursor as I have with Opus 4.6 in Claude Code.
Opus is obviously the better model, but Cursor's "harness" is doing so much heavy lifting in terms of just magically supplying the broader context the model needs to understand the ramifications of its edits.
One thing I enjoy about Cursor and Codex mac apps is the embedded preview window. I know it's not as hardcore as the terminal/tmux but it's hella convenient. But Cursor bugs me with the opacity around what model I'm using. It seems deliberately to be routing requests based on its perceived complexity. What draws you to codex vs cursor?
I think removing Claude Code from the $20 tier is a terrible idea, I never would've gone from nothing right into the $100/200 tier. The $20 plan let me get my feet wet and see how good it could be, and in less than a week I was on the $100 plan.
I think they need to at least have a 1 month introductory rate for the max plan at $20, or devs that decide to try out agentic coding just won't go to Anthropic.
That leads to downstream impacts, like when a company is deciding which AI coding tools to provide and the feedback management hears everyone is already used to (e.x.) Codex, then Anthropic starts losing the enterprise side of things.
They're not losing anything. They have much more demand than they could ever fulfill to care anymore about promotional or subsidized user groups.
Until magically all their demand vaporizes.
I suspect a lot of people are like me. They got into this at the $20/month level individually to check things out. I'm not stressing things out, so I haven't moved up, but the moment I bump into a limit, I'll pull the trigger by default. Until then, I'm the sleeping dog, and you should let me lie.
Well, Anthropic decided to kick me. Now, I'm investing the time to figure out how to use the "open" and "Chinese" models assuming that Anthropic is about to screw me. Once I switch, Anthropic is going to have to demonstrate significant improvements over what I'm now using to get me to even consider them again.
I don't think they need the $20/month users when there are some who use over $1000 in tokens per day.
I know of 4 companies that are already starting to stomp down on the AI whales using the "$1000 per day". There was the cost of the entire AI usage of the company and then there was the cost of about a half-dozen people who dwarfed it individually.
So, we've established a hard upper ceiling for what AI can extract per user at roughly $100K and more realistically at $10K per year. Basically, if using the AI costs the same as a human salary, it's going to get pushback. I mean, the whole point was to get rid of those pesky human salaries, after all.
So, there are about 2 million-ish software jobs in the US? It's more than 1 million but a far cry from 10 million. So that pencils in at $20 billion in the US per year total? That means that if an AI company literally won all the US software programmers, it would be worth max $200 billion to be bought out (10x revenue).
Now how much investment have the AI companies taken? Yeah, roughly that. And investors are going to want quite a bit more than that back.
Even if they had zero delivery costs, the AI companies are cooked long term. The moment your number bumps into "All the X in the US/World", you've got a problem.
Short term? Greater fool theory applies. And there appear to be a lot of them.
And all this is before we start getting into people exploring the open models. Most people were like me; we started on something like Claude and just stayed put because it was straightforward. Now that we've been kicked, we'll start looking at the other options.
I agree. Why would they not keep the $20 plan as a gateway drug?
LLM monsters are deeply unprofitable, going by the industry hearsay (which is the only thing we have, given ultra secrecy of the LLM corporations). The only two LLM companies which disclosed their finances without lies, were two Chinese corporations and they, unsurprisingly, were deeply in red.
Remember the old saying about boiling a frog? LLM corporations need to make most of their users pay hundreds per month, asap. This is Anthropic increasing temperature regulator under the pot just a tiny little bit. Not the first and not the last time.
Their price point goal is a SWE salary.
I assume the Chinese corporations can operate in the red forever and be subsidized by the Chinese government.
That's true of any company in any country. If you can convince the government that your company is sufficiently important, you can get subsidized.
My point was that the AI companies in China have already convinced the government to subsidize them.
Good for them.
Tax breaks, fee reductions/waivers, direct monetary incentives, and shielding from "unfavorable" regulation -whether local, state, or national- are all subsidies. Hell, depending on the particulars, government contracts can be subsidies... there's more than one government engineering project out there that could be reasonably referred to as a "jobs program for PhDs", and still more that are corporate handouts.
Every business believed by its "home" government to be sufficiently important gets subsidies when it asks for them... regardless of what nation houses that government. If your claim is that the major players in the "AI" industry aren't getting subsidies from local, state, and national governments in the US, then my claim is that you are lying.
Matches my experience very well. All the goodwill earned from taking a stand against the DoD seemingly forgotten in a month. Coincidentally, I canceled my pro subscription and got set up with OpenCode and OpenRouter last night.
Got any good pointers to documentation for making the transition? I'd like to pull the trigger for OpenCode and OpenRouter as well.
I just installed it and continued about my business. I don't have a carefully tuned claude code setup. It has some skills, some CLAUDE.md files, and not much else.
I did also migrate an app from Claude SDK to Pydantic AI to get off claude API pricing. https://pydantic.dev/docs/ai/overview/
> I can't understand removing Claude Code from $20
Not according to their webpage: "Claude Code is included in your Pro plan. Perfect for short coding sprints in small codebases with access to both Sonnet 4.6 and Opus 4.7." [1]
[1]: https://claude.com/product/claude-code
There are clear contradictions across their marketing site. As others have pointed out, it's being removed from some help articles and the pricing chart now shows it revoked. Confusing signals, but they seem to be changing all pages in this direction and haven't updated that one yet.
See https://news.ycombinator.com/item?id=47854478
what happened to agentic superintelligence based development?
This is superintelligence. The mixed signals are tested to increase their revenues. Superintelligent AIs wouldn't be honest.
> I can't understand removing Claude Code from $20. I'm interested to see whether this is confirmed or not.
Anthropic bleeds money per user. No matter if it's the $20 or $200 plan, every Claude Code user is unprofitable.
The only way to not bleed money is to eventually move everyone to API pricing. Hiring a personal senior engineer will likely be cheaper.
> I quickly started hitting limits developing several apps at the same time
> I'm a career engineer
I'm trying really hard here to be nice, but what the hell are you doing? Are you vibe coding multiple apps in parallel and calling it engineering?
Is it like those people who eat 2-3x the amount of meat to ensure they offset the positive impact 1-2 vegans are having? :D
I've noticed that there are some people who feel that their claude instance could be working on churning out multiple apps, and therefore if it isn't they are in some sense falling behind. it's the illusion of productivity raised to the level of a minor addiction.
I had a similar ride, but disagree with your conclusion. Opus 4.7 is so incredibly powerful from my experience, that nothing else really matters and I think at Anthropic they know it. People will pay a lot for access to this model.
>Opus 4.7 is so incredibly powerful from my experience,
I'm not challenging your opinion, but this is an outlier in the general current public opinion about it.
This is one of the most civil disagreements I've ever seen on the internet and I intend to start using this myself
Yea, I've seen a lot of whining online, because its more expensive, but from the interactions I've had I'd say, that it's well worth it. To me it feels like another step change, similar to when 4.5 was introduced. Definitely a different beast.
EDIT: it is also surprising to me that everyone seems to believe the people at Anthropic are simply incompetent and recklessly risking their good reputation, while very few consider the possible good reasons they might have for taking such drastic measures. And I don't think it's because of financial pressures in their case
I can’t say I’ve used it extensively enough to draw a conclusion, but it did seem similar to GPT 5.4 in Codex.
When I threw it at a difficult issue in an iOS app, it like GPT came up with wrongly guessed explanations. It only found the issue after I had it instrument the app and add extensive logs. Usually GPT 5.4 is the same.
Only that with GPT 5.4 it’s at least included in my subscription, while sending 3-4 messages to Opus 4.7 for this blew through my $20 plan limits and consumed $10 of extra usage on top. At that point I can’t help but bring up how much more expensive it is.
> Only that with GPT 5.4 it’s at least included in my subscription, while sending 3-4 messages to Opus 4.7 for this blew through my $20 plan limits and consumed $10 of extra usage on top. At that point I can’t help but bring up how much more expensive it is.
Rest assured OpenAI won’t want to leave that kind of money on the table…
There’s also still Google with their TPUs, xAI has some large models in the works, not to mention China.
With that much competition and ongoing improvements, I don’t have such a pessimistic view on future usage limits and cost.
Having too much usage and not enough GPUs is a form of financial pressure, no? since you want to replace your less profitable customers with more profitable customers
I personally prefer 4.7 to 4.6 or previous model.
I've had completely the opposite experience. I've asked for it to research things and it's just told me to "paste xyz into google". Just now I revisited a chat that's 5 days old and asked it to check again (because what I was looking for might have changed), and it said "no".
It's funny how experiences can be so different. I wonder if this comes down to context. My interactions so far were fairly high-level and in some cases it having a strong opinion was actually super beneficial to the outcome. To me it seemed opinionated, but in a very good way. I can see how this could backfire though and have heard similar reports.
Curious to know what plan you're on? I was on the max 5x plan, but downgraded to pro a few days before the opus 4.7 release.
I'm on max 5x
Incredible, powerful, but I couldn't believe how fast I hit the limits compared to how it was with Opus 4.6. They removed Opus 4.6 completely from CC. I would prefer it with the previous limits.
That's not how you keep your customers. None of these agents have a moat, I moved away from Cursor when they started doing what Anthropic is doing now, and never went back even when I was a paying customer since the start.
You can just use the model parameter to bring it back
Opus 4.7 may be incredible but for how long? And they may have Mythos but I feel like they will put it out if pressed too much by their competitors. And again for how long will they keep the advantage?
At the speed everything is advancing I don’t think it’s such an advantage. They all catch each other up pretty fast. That’s why I prefer to pay Cursors and have access to all of them instead of being lock to a single one (even if that means to lose some discounted credits). If they opened Mythos today at a good price that would be something but that’s not the case and it won’t happen.
they need the devs on board for that to matter, i can get whatever i want done with lesser models already. It is quite literally about just who is not gonna give me the shittiest experience, and at anthropic it sure seems they are determined to annoy everyone since they started gaining in popularity.
[flagged]
FWIW their "Fin" agent on their web page tells me: "Claude Code is not included in the standard Pro plan. It's only available with premium seats on Team and Enterprise plans. However, Pro plans do include improved coding capabilities through our GitHub integration, which lets you connect code repositories directly to Claude. You also get access to code execution and file creation features in the Claude apps. For the full Claude Code terminal experience, you'd need to upgrade to a Team or Enterprise plan with premium seats"
Even odder. I asked: Are you sure?
Yes, confirmed directly from Anthropic's website. Claude Code is included in the Pro plan, described as perfect for short coding sprints in small codebases with access to both Sonnet 4.6 and Opus 4.7. Claude
You said: Is claude code included with the pro subscription Is claude code included with the pro subscription
3:11 PM Claude responded: Yes, Claude Code is included with the Pro plan ($20/month). Yes, Claude Code is included with the Pro plan ($20/month). You can use it in the terminal, desktop app, VS Code, JetBrains, and on the web.
The main caveat is that Claude Code draws from the same usage limits as your regular Claude chat — so heavy coding sessions can eat into your quota faster. If you find yourself hitting limits often, the Max plans ($100–$200/month) offer significantly more headroom.
Thats a head scratcher:)
The $20/month, which is now $17, should give you a hint? Models have a knowledge cut-off and will not reflect up to date information unless you trigger a web search.
That's just a billing trick.
>$17 Per month with annual subscription discount ($200 billed up front). $20 if billed monthly.
Do you understand how LLM's work and that they are always behind in their knowledge? Unless Claude does a network call to check its own website, it will give you outdated information. Its a prediction model, its not magic.
Claude often doesn't know the truth about Claude Code etc. lol
I got down voted by 4 people just for pointing that out. lol
What a way to ruin goodwill with the very community they are trying to court. I am on a Pro subscription to use with Claude Code, but it sounds like the days of using it are numbered. I guess I will be trying the latest offering from OpenAI and Google tomorrow and if they are satisfactory I might just switch. Moreover, I have been recommending Anthropic's API solutions up to now to friends and clients. Based on this dumb move I will be now starting with this anecdote and then giving a very hedged recommendation.
Realistically the future of all this is that open models become good enough that LLM as a service becomes a commodity with a race to the bottom in terms of cost. Given where we are today I can easily see open weight models in 2-3 years making Anthropic and OpenAI irrelevant for everyday development work (I justify this like so: if my coding agent is 10x smarter than I am, how would I understand if it did all the right things? I want someone of roughly my intelligence for coding. I can see use cases for like independent pharma work or some such where supergenius level intelligence is justified, but for coding ability for mere mortals to reason about the code is probably more important).
> the very community they are trying to court
After all, we may be a just a data source and not their intended demographic all along.
The valuation is obviously based on the premise of their capturing the white collar economy. OpenAI's charter says so openly. And Chinese robots will come for blue workers next.
The economy, not the workers :) It feels like pretty soon white collar workers will be in a “You have nothing to lose but your chains” situation. Except we are not as fit as the proletariat of the past.
In my experience, Codex is better than Claude Code in every way and GPT-5.4 is on par or better than Opus 4.6 at every coding task I ask of it.
You're really not going to miss CC. And OpenAI actually had some foresight to invest massively in compute so they don't run into usage and rate limits like Anthropic does constantly. I couldn't even use CC for more than a couple complex tasks before I was out of extra usage or session usage. It was a maddening productivity killer and I just switched to Codex full time.
> I guess I will be trying the latest offering from OpenAI and Google tomorrow and if they are satisfactory I might just switch.
If Anthropic’s move is confirmed, my guess is other coding agents providers might end up making similar moves
This is the definition of cartel
Gpt xhigh isn't that bad..
Kimi K2.6 is supposedly good: https://www.kimi.com/blog/kimi-k2-6
I am on Google's $20/month plan, and I usually get about three half-hour coding sessions a week with AntiGravity using the Claude models. The limit using Gemini Pro models is much higher. I am retired so Google's $20 plan is sufficient for me, but I understand that people who are still working would need higher limits.
I am also on a $10/month plan with Nous Research for supplying open models for their open source Hermes Agent. I run Hermes inside a container, on a dedicated VPS as a coding agent for complex tasks and so far I find the $10/month plan is enough for about five to ten major tasks a month. I think it is also a good deal.
If I could get the equivalent of GPT-4 running locally, that would cover like 95% of what I need an LLM for. Tweak this dockerfile, gimme a bash script. I guess the context probably isn’t sufficient for the agent stuff, but I’m sure more context-efficient harnesses will be coming down the line
I have an old Mac Mini with 32G of integrated RAM, and the following works for me for small local code changes:
ollama launch claude --model qwen3.6:35b-a3b-nvfp4
In addition to not having an integrated web search tool, one drawback is that it runs more slowly than using cloud servers. I find myself asking for a code or documentation change, and then spending two minutes on my deck getting fresh air waiting for a slower response. When using a fast cloud service I can be a coding slave, glued to my computer. Still, I like running local when I can!
[dead]
gpt 5.4 has been performing great in my harness.
I have codex and Gemini for spill over, they work good.
It would signal quite a fundamental pivot if their "Pro" plan excludes coding but supports personal productivity (Cowork). Quite surprising given most people attribute Anthropic's success to their elevation of coding above everything else. To have casual users locked out of that would be a major hit you would think.
Makes me curious about the internal thinking. One theory being they are in a capacity crisis and knocking Pro users off Claude Code is an emergency brake getting pulled. But an opposite theory is it's a revenue move and they think they have the lock in to pull it off. Especially if they are building up to IPO.
Interestingly the Team subscription which is still $20/month/seat still includes Claude Code. But you need minimum 5 seats. So it could be a way to force people off individual plans and into enterprise plans where possibly things scale better for them, especially IPO/wise. When one user wants it in a company, probably they go buy 5 seats.
I have to assume they're compute constrained and thus need to either raise prices or cut their lowest-margin products (which amounts to more or less the same thing, but with different optics), or turn away new users.
My assumption is that people are able to very easily saturate Pro with Claude Code and therefore even though the quotas are lower (more than proportionally) the utilization of those quotas is higher enough that Pro is less profitable.
I think there is a definite possibility that they aren't compute constrained, but rather trying to improve a sorry cash flow situation before IPO.
Of course, I don't have real insight into available compute, but the vibe slope seems to have dropped a bit, at the same time as new GPUs are being shoved into datacenters as fast as possible.
Their enterprise API customers are literally competing to see who can throw the most money at Anthropic. Anthropic has very little reason to focus on a $20/month user, and with their current momentum (especially since enterprise deals are long-lived) they could remove Claude Code from the Pro plan without any revenue hit. In fact, it may be a huge revenue boost given the strength of the Anthropic brand.
If that's the case, what will happen after IPO? Will they become good again?
I just switched from the $10 Copilot subscription to a $20 Claude subscription to get general AI and coding in one bill. I guess I'll try out GPT Codex.
gpt allows you to wire their models into other CLI tools, I'm advising everyone I know to lean that direction. Not trying to become hostage to something like claude's ecosystem for the rest of my development career.
They will eventually converge --- it's only a matter of time.
It's possible that Anthropic sees that the loss of $20/mo customers could be offset by the customers purchasing the $100/mo plan
This is a market where hyper growth is important
Loss of customers is the wrong direction
Their supply of being able to serve the models is hardcore rationed by their ability to scale up datacenters and GPUs.
They really don't need to subsidize unprofitable customers at this point when there is a line out the door to pay thousands of dollars a month per user, that are revolting because they aren't actually being able to get reliable uptime.
They have all the growth they need for now, they really don't need the cheap users.
Why would you even want a Claude subscription if not for Claude Code? Anthropic is obviously the best for programming, but probably nowhere else. Seems like a good way to onboard people to the Claude Code experience...everyone who's working seriously with it needs Opus, anyway. But, maybe that's the rub, if the Pro plan includes no Opus usage (which I think has always been the case), you might have a worse impression of Claude Code. Codex 5.4 is better than Sonnet, but not better than Opus.
I dunno, I'm no business genius, but I think we're starting to see these companies try to find ways to make money instead of losing it.
The pro plan does include Opus usage. I've noticed the limits on the web client are a bit higher than through CC, but probably more because of the increased token usage of agentic coding in general.
Claude web is actually pretty good for dealing with random projects outside of code. I have a Home Assistant MCP server [1] behind a Cloudflare tunnel exposed to it that makes maintaining automations a lot easier.
[1] https://github.com/homeassistant-ai/ha-mcp
On LMArena, Claude Opus is ranked as the best at everything except image and video generation, which it does not support. That may be inaccurate, but it's plausible
> Anthropic is obviously the best for programming
Not at all: https://i.imgur.com/jYawPDY.png
Codex has been way better for my use case (reviewing existing Godot+GDScript code)
People subscribed to chatgpt before there was codex. Why wouldnt a Claude subscription stand on its own without Claude Code? In fact it’s probably a smart move for Anthropic to split it out.
Because Claude Code provides me more value than simply a chat agent. Claude Code is the only thing I pay for. I can do chat things with the free tier across all the LLMs.
Opus 4.5 was part of the pro plan and 4.6 too
Wow, downvoted for stating facts
I have been using https://claude.ai and, initially, it was good, but, unfortunately, it keeps getting worse. I had it search for contact information for a certain public entity, and in Claude's response, all emails were being replaced with [email protected] or something like that. They also added an absolutely horrendous automatic markdown in the text input, so now you can't even properly enter your prompt. It actively gets in my way and prevents me from typing what I want. Fuck you Anthropic.
Next they’ll slowly reduce how much CC usage you can get out of the $100 Max plan, then introduce a new $300 “Max Plus” plan with “40x” usage.
“You asked, and we listened: Introducing Max Plus, our biggest plan yet, designed for those…” blah blah
They already effectively halved it with the introduction of Opus 4.7 and the new tokenizer that basically gives you about half as much usage for the same price.. Convenient to price based on tokens, and leave what a token is a moving target..
Source: I could easily work with the anthropic quotas before. Now I run into a wall constantly.
I haven't changed the way I work, I've become more conscious of context sizes than before if anything.
Still run out of quota constantly with 4.7.
Source?
Look up independent tests of the new tokenizer for opus 4.7. I don’t think it was double, more like 140%ish of Opus 4.6.
Yup, and Uber rides that used to be $5 are now $12. Because they stopped burning investor money and taking losses on each ride.
combined with 30-50% token inflation too. hmm.
I have a Claude Pro tier subscription; Claude Code, as of right now, is still functional for me. If Anthropic does boot Pro-tier users off Claude Code, I will be cancelling my subscription.
Indeed. Codex on $20/m is incredibly usable. Lots of value. My anthropic subscription keeps being worth less and less.
> Codex on $20/m is incredibly usable.
And how long do you think that will last if A\ does this?
I'd almost bet that OpenAI will do the same thing with its plans as soon as Anthropic makes the first move. It's just a matter of who blinks first.
I paid for the annual Pro plan in January...I know this mentions new users right now, but is there a chance they just take Code away?!
They would probably grandfather existing users in for at least a year or something, you have to imagine. Even if this "test" goes very well and points to removal
This test makes perfect sense with their actions the last few weeks, they think they've done enough to transition into the general public and away from devs and our goodwill no longer is something they should be concerned with.
Its funny that openai, who in my eyes went for the general public rather than devs initially, seems to be semi pivoting and catching all the fallout from anthropic's recent behavior.
It is a massive bummer, up until those few weeks ago, i was hard pulling for anthropic for quite some time, now i just dont care and hope something dope emerges quickly that signals i wont ever have to consider either of them.
I think they will try to get into the enterprise as soon as possible - looking out for big purchases - figma, atlassian, asana, monday, etc...
Yeah, at 100$ or 200$ a month my expectations would raise (and tolerance to errors go to zero) as we are going into enterprise level pricing.
I can't find an announcement yet, however the pricing page now shows it's not included, and various support articles have removed any mention of the the Pro plan including access to claude code.
See [1] and [2] for an example of a support article that's had claude code removed as a Pro feature.
I guess this is the beginning of the end for subsidised model access, at least from Anthropic.
[1] https://support.claude.com/en/articles/8325606-what-is-the-p... [2] https://web.archive.org/web/20260420065828/https://support.c...
It’s now being explicitly shown as unavailable on the pro plan, scroll down to the comparison matrix.
And this page as well, showing how to use Claude Code on the Max plan: https://support.claude.com/en/articles/11145838-using-claude...
Claude Code is a freely downloadable CLI Agent. Why would they not let you use that anymore?
Because 'claude -p' is a backdoor for any third party client to use subsidized token pricing, and they've flipped course again to decide they do want the OpenClaw type users as long as they're on the $100 plan.
> Because 'claude -p' is a backdoor for any third party client to use subsidized token pricing
Wasn't this obvious from day 1 though? Can't see how anyone could've missed that.
It was obvious, which is why their earlier decision to attempt to restrict 3p access was hugely unpopular. They're now trying to walk that back but on the condition that those users need to be on the higher plan.
Now they're hurting their popularity with those who actually don't mind using Claude Code. I've been quietly swallowing a lot of performance degradation over the past few weeks as I get the resource crunch, but I'm definitely not going back to copy-pasting between browser and editor. And I have no intention of upgrading to Max or doing per token usage.
It seems they were losing money on pro subscriptions with code so if people stop using them it's actually ok for them.
They have now moved to be enterprise providers and don't need the cheap pro users as loss leaders anymore.
The question isn't whether anyone could have missed it, but whether Claude Code has release gates that allow people to require that obvious problems should be resolved. From their release velocity it's pretty clear that they do not.
You have to login and authorize it? It costs money to process tokens.
Because including it in a plan results in un-metered usage?
My usage of Claude Code in the pro plan is definitely metered. Every couple hours I have to wait an hour or two and the last few weeks I've hit my weekly limit on Wednesday.
5 minutes ago, I was seeing the old version of the page in which Claude Pro included Claude Code. I refreshed and now see that it does not.
This is a risky move. I might have paid $20/month for my personal projects but the Max subscription is a bit steep.
Now I'm going to learn more about local models. I don't need to be as good as a frontier model. Good enough and free from all this drama is a win for me
Same, I'm currently unemployed and the $20 help me to initiate many small projects. The recent taxing in the token made me start testing local models on my machine. Tho, Claude works better for the front end part imo.
If you are looking at open models, check out Pi, as its very extensible and comes with a sane default. Maybe even roll your own.
Most harnesses (claude, codex, opencode etc.) assume that you use a cloud model. There’s no sense of optimization or finer control.
I've been running pi.dev + codex (GPT-5.4) for a long time for my workhorse stuff.
Actually tried caveman mode yesterday and it made everything SO MUCH BETTER. GPT-5.4 has a habit of being extremely verbose to a ridiculous degree, it's like it's writing a report for a CTO or something and padding everything as much as possible to sound smart.
With caveman it just gives me lists of stuff in a compact format. Perfect.
let's be clear(er) here that you like caveman's format and output, so that's one value of "better". I've seen at least a few (n < 10, so there's that) tests on token use and for THAT value of "better" it's not much different.
The token savings with caveman is completely secondary to me to be honest.
What I appreciate is the terse answers it gives to me. I don't need a padded essay on what the issue is or what was fixed. Give me the bullet points.
"free of this drama" and free is great option for companies, of course most use API billing but let's not forget that there are places that budget is limited and being good enough is just perfect.
the cursor 20$ a month plan has been working great for me. You can use most models, and unlimited use of composer 2, which is surprisingly good
It's a volatile space.
Cursor was just acquired by SpaceX, so let's see what happens.
Wait what
Front page right now.
I would love someone to play devil's advocate against this perspective:
While these tools stand to enable the democratization of productive capability in software engineering and other tasks (creating a renaissance for solopreneurs, let's say), what seems more likely to actually happen is that entrenched capital will become the only player with real access to this "knowledge as a utility" (was it Altman who called it that?).
We already see this playing out in two fronts: 1) the gradual reduction of services and 2) the DRAM market, where local-first tools (i.e., potential disruptors of the emerging "knowledge monopoly" created by the big AI firms) are being stifled by supply shortages. How many promising small-to-medium-sized competitors are being snuffed out of existence (or never starting) due to the insanity of the DRAM/storage/CPU (soon) markets?
The currently-subsidized access that we have to the big Opus-like models will, in parallel, be gradually be taken away until only the big players can afford it. And in the end what we will have is hyper-productive skeleton crews at a few consolidated firms performing (or selling expensive access to) basically all of the knowledge labor for society, with very little potential for disruption due to the hardware and "knowledge" scarcity engineered (in part, maybe) by this monopoly.
Not necessarily a closely held belief – just a hunch – which is why I want to see what parts of the picture I might be missing.
Devils advocate here - pro and max tier customers for all the major inference providers are loss leaders from the data we have been able to figure out, and reverse engineer. They are effectively a marketing exercise.
The real profitability is selling tokens to enterprise, and enterprise demand is growing so fast that they are short on the total amount of tokens they can generate per minute, and are prioritising rationally - enterprise gets a better experience - instead of optimizing for their lowest paying (and most loss leading) customers.
We are in a hardware crunch right now but that won't be forever, and eventually (likely 2028) we will get experiences like we got in January from pro-sumer accounts again.
Not only because of cost. Mythos has only been released to some of the big tech players because it's "too dangerous" [0] for us little people.
It's easy to see this becoming a permanent position; the latest models and smarts are reserved for establishment members only, the riff-raff get the cast-offs. So the establishment is preserved and the status quo protected.
[0] I'm putting scare/irony quotes around this, but if the reporting is accurate, there is something to this; we built the internet on string and duct tape, it's not hard to see how a very smart AI could cut it to ribbons.
Brother, you are falling for marketing speak.
In periods of massive inflation, only the most wealthy survive.
But there's competition out there -- the open-source chinese models. In their current form, I assume that will turn off many people but new models -- based on those -- are likely to appear. Also, OAI and Google will release new models and pick up the lost customers.
With GLM and Kimi getting better and better, with both still providing low-cost coding plans with higher quotas, and with how trivial it is to switch to them even within Claude CLI, I'm not sure what makes Anthropic think their users would rather pay 5x than switch to the competition.
Never mind GLM and Kimi, even GPT 5.4 offers a great plan for $20/mo. Even if it gets increased rate limits after May, it's still quite likely that casual users won't be hitting them on a regular basis.
openai is too toxic for many
And I'm sure their bean counters are watching this VERY closely; if it doesn't sink A\ outright, there's no reason they won't up their prices too.
I signed up for a 3 month plan with Z.ai so I could try out GLM-5.1. That was a few weeks ago when it was still $27 for 3 months, now it's $48 for 3 months. I hit limits at least as much as I do with Claude. I hit a weekly limit at one point and it said I wasn't going be able to get access again for 6 days so I must've somehow hit their weekly limit on day one of the week. And that after several timeouts.
The GLM coding plan price increased dramatically
It is left unsaid, but throughput is also terrible.
Why is management at Anthropic trying so hard to ruin their reputation with developers? I missed the OpenClaw hype but it was something that kept me excited about my yearly subscription.
It makes no sense to do one of the higher tier plans unless they are directly generating you money.
They care about developers from companies that are on their team/enterprise plans or using bedrock.
Individual users barely matter. That's probably also the same group that decides to switch to Codex/Kimi/[whatever the hottest agent on any given day] on a whim, which Anthropic doesn't necessarily want to do business with.
feel like its beyond optimistic on their part, just starting to hear their name be blended with companies desires on job listings, and they are destroying the goodwill of the devs who surely are the main reason their name has landed there. They aren't dug in like a microsoft, maybe they get some staying power for nocode people who feel trapped, but im done with their nonsense already and won't recommend them anywhere. Other stuff is good enough already to match.
> Individual users barely matter.
Individuals are the ones that push for new tools at work though.
A fraction of them do. Many just use whatever the employer provides to get their job done. HN users only represent a small sample of the overall software developers which is nowhere nearly enthusiastic about new things.
Source: what I witnessed at my company
In the end, companies are made of individuals. In my previous company, I'm one of the individual who advocate for Claude Code adoption. Amongst my circle too, most who have the authority to make an impact on the AI direction. Safe to say I no longer suggest Anthropic to anyone anymore.
This is facts. Most of the devs at the huge companies Anthropic and the like try to court are using Copilot since it comes with MS365.
they don't care about their reputation with devs, they care about their reputation with people that can write them big fat checks
At my company, devs were the ones pushing for the Claude subscription. Left to management, we would have only had GitHub Copilot – we already have an existing relationship with them and the tool is good enough.
If Anthropic is intent on losing the goodwill of the devs, they might not be happy with the consequences. Their product is quite commoditized at this point – the latest GPT, Gemini or GLM is just as good for most enterprise tasks.
Exactly the same for me, and I'm now ... "worried" isn't quite right, but you know what I mean, that they will back out. But to what? We had copilot before which was cheaper, and worked reasonably well with the A\ models, but I'm not even sure those will be there (Opus is no longer on my cheapest paid Copilot sub at home), and I've no doubt OpenAI will jack up their prices soon enough and/or do the "exclusivity" thing so they can only be accessed by their own clients.
Where I work. Medium size, base in Europe company. It is paying over 1800 per dev in AI tools. Home users stand no chance.
Per what time frame? Day, week, month, or year?
> Why is management at Anthropic trying so hard to ruin their reputation with developers?
They follow the money, like every other company.
I suspect their ICP is changing from developers to enterprise decision makers, completely different personas
Also note that they are letting OpenClaw be used again with `claude -p`, so a partial reversal
And I thought MS was confused one on how to do pricing and business decisions.
Last I saw they have 21 products or services named Copilot. I think they still win the confusion prize.
We should have "The Copilot Award goes to..." Every year for the most confusing product name/lineup
I love it. It’ll be doubly funny when the Xbox marketing team wins the Copilot Confusion Award.
https://news.ycombinator.com/item?id=47853799 for the curious
They got contracts from Enterprises now
https://x.com/TheAmolAvasare/status/2046724846604963924
(Head of Growth @AnthropicAI)
> When we launched Max a year ago, it didn't include Claude Code, Cowork didn't exist, and agents that run for hours weren't a thing. Max was designed for heavy chat usage, that's it.
Is there a wager that this is 100% foreshadowing Claude Code will be removed from the $100-200/month Max plans soon and go to something like API-only? Or only available on like a new $500-1,000/month plan? Restrict the $100-200/month ones to Claude.ai (website) or Claude desktop app only?
Either way, doesn't seem good to say it's a small test and then start justifying it in this direction.
Do they have a substantial userbase for this outside of claude code? The only two use cases for LLMs that seem to have significant traction are programming, and erotic roleplay lol. If they stop catering to devs, who is their market?
FWIW, I just heard this guy (Head of Growth) on Lenny's podcast a few days ago and one of the things he explicitly mentioned was creating intentional friction for growth. This seems to be one of those tests.
That would be a fantastic way to kill the majority of engagement with the community of users. There are very few who can afford that.
Makes sense.
It is over for the little guy - home enthusiasts and vibe coders. Too many of them saturating resources for Max users.
IF you cannot afford few hundred dollars subscription go out and breathe fresh air. But if you can, watch where the ball is rolling - few thousand dollars subscriptions and even less programmers.
Hear HN tell of it, Claude pays for itself 3× over.
Something tells me congitively it's making us misjudge how productive it's making us.
It's clearly massively increasing output, but did the market already soak up all that productivity and now it's not compensated?
If your salary is 50k. And Claude makes you 2x as productive, why aren't you earning 100k?
Why is it anyone can't afford $200/mo if it's truely increasing worker productivity?
There seems to be a paradox here.
Personally I switched to Z.ai and GLM quite some time ago. I've not noticed any decrease in quality or quantity of my work.
Agree about psychological impact outpacing likely actual impact, but that’s a relatively temporary phenomena as we are all adapting to the new way things work.
Productivity wise employment is far more than code production productivity in a vacuum, and productivity gains are rarely captured by employees (see famous chart on worker productivity where that correlation changed around 1970). I wouldn’t expect to see much in the next 1-2 years besides noticing effective teams increasing velocity of features.
I think people in forums like complaining about things and aren’t representative of the broader set of people who are just using the tools, so no real paradox. For vast majority of tech jobs, $200/mo is still an absolute steal in terms of what these tools offer. Only the dullest of companies would not realize this.
Fwiw in the 80s-90s computers also didn’t really register in productivity metrics. Qualitative changes occur long before accurate measurement catches up.
Because most people work for someone else and don't decide their own salaries. It's not doubling productivity, but even a 10-20% boost to productivity for a team of engineers means that, as a business, even $1k per month per seat is perfectly acceptable. For consumers and hobbyists that basically kills access.
yeah the more people who use it means less competitive edge you have. Benefits get devalued. And you're back to square one.
> Personally I switched to Z.ai and GLM quite some time ago. I've not noticed any decrease in quality or quantity of my work.
> Something tells me congitively it's making us misjudge how productive it's making us.
This could be happening to you, too.
Vscode agent mode and github copilot can use Claude models and has feature parity with the .md customization for agents prompts skills etc.
Not too expensive
They slapped a 7.5x “promotional” multiplier on Opus 4.7 and they are removing Opus 4.6 in short order.
I heard they disabled signups for non-business accounts too.
Best forget about using Claude Opus models in Copilot.
> Best forget about using Claude Opus models in Copilot.
I noticed this morning that Opus isn't even one of the models in the `/model` command in Copilot. Highest I can get (on the paid, but least expensive) tier is Sonnet 4.6. I'm pretty sure Opus was allowed recently, but not now.
Yeha not thrilled about that.
Looks like you gotta build your own agent harness and self host or use aws bedrock for "sovereignty"
Indeed; especially since I paid for a sub with some expectations, and those are being changed out from under me.
I heard they offer a full refund for this month if you are understandably unhappy with these changes.
Truly makes no sense. I pay for the $200/month plan and end up using about $3k/month worth of API costs. I imagine that the only reason they haven’t cut me off is because my habits serve as good training data for them.
Guess they’ve decided to move in the direction of allocating compute primarily to power users and enterprise.
But power users are not a sticky customer base. I just bought the ChatGPT Pro plan and would immediately switch over if the model performance is better and/or I get more compute.
Or the API is overpriced. The concept of charging per tokens does not map well to the actual costs an AI company has.
Odd, everyone was insisting this would "democratize" programming though.
Guess it democratizes it if you have money, huh?
Good old "one dollar, one vote" democracy!
Most commercial websites, especially tech, run multiple A/B experiments in parallel to optimize signup funnels and tier conversions. Even so, there should always be a source of truth for people. We should have some way of verifying what we get for what we pay. There are laws about this as well.
This guy's casual and crass response is a sign of disrespect for customers. Unfortunately, that is pervasive in the industry. The bubbles these teams work in are corrosive to empathy and real world impact.
I think there is more here: Anthropic's whole market positioning is based on trust. It's literally their reason for being.
The Claude constitution has a major section about not being deceptive. Now this is GTM, not the model, but there is clearly a coherence problem here... and if anyone should realize the important of their market positioning it's GTM.
Agreed that it’s 100% marketing. In some ways Anthropic is more of a for-profit corporation than OpenAI which is at least partially owned by a non-profit.
Claude has become practically unusable for Pro users in the past few days. The Opus 4.7 blew through an entire 5 hour limit in one question and didn’t even finish answering it. Zero value delivered.
Opus 4.6 is giving 2, maybe 3 questions before blowing through the Pro 5 hour limit as well. We are forced to use Sonnet which makes the same mistakes over and over and then to start trying with other companies. To make matters worse, it reuses old code as we try to survive between credit expiry so it re-introduced issues into the code with the limited credits, that we had already fixed on our own and with other models.
Anthropic in just a few days has gotten me to try GLM 5.1, the new Kimi, and back to OpenAI. OpenAI also seems to introduce new bugs without being carefully micromanaged. The advantage Claude has is that the models are more careful and can refactor code instead of leading to bloat as they go. But the throttling happening now is breaking things and making the entire subscription unusable. I really hope they fix it soon.
I'm starting to think I've been A/B tested, because this was my experience for almost a year with Claude ever since I tried it for coding. Meanwhile, my coworkers seemed to be able to use it for long periods of time without getting rate limited.
One interesting variable is that I'm located in Vietnam while my coworkers are located in Norway and Europe.
To work around this issue I used Claude for coding with a Copilot subscription which was much cheaper and had virtually no rate limiting.
Copilot gives you some set amount of credits each month, but you can also pay as you go if you run out of credit which is much better than the 5 hour window crap claude code would give me.
The only opus model available now on copilot for some reason is 4.7 and it costs 7.5x tokens, while everything else is 1x, 0.33x or free.
But I switched to using GPT 5.4 medium for a month or so which I find very reasonable.
My personal LLM coding stack is now OpenCode, Claude Sonnet for ideation on spec with OpenWhispr for voice-to-text, GLM-5.1 for the orchestrating loop, GLM-4.7 for coding, and DeepSeek R1 for review and validation. It works much, much better than the Claude Code setup I have at work for substantially less money to boot.
At this rate I fully anticipate being able to run a comparable stack on a 128GB Mac Studio using quants of newer-generation distilled OSS models in a year or two. Being able to ramble to a computer for an hour about features and technical philosophy then have it build a nearly-working app for $50 is an exciting feeling. There's still a long tail of productionization and fixing what the model didn't adhere to but it's still incredible.
Im locked in for a year of claude pro, I encountered the same issues as you a couple weeks ago, Id get like one solid plan done and really really hope it was a 1 shot because that was legit all i was gonna get out of it for those 5 hours, and it would be ~10% of weekly usage to really make me feel scared to hit send.
I got the 20$ gpt tier, and now i just use claude to craft MD plan docs instead, and then i hand them off to gpt 5.4 and it has been working great. can do about 4x as much work or so based on my feelings(not accurate). if i have just small simple stuff to do i might still fire those off with sonnet and that seems plenty viable, but as soon as its an opus tier task i swap to this workflow.
Little annoying as now im kinda trying to manage a .claude/ and an .opencode/ folder but i kinda just have the .opencode/ stuff reference the .claude/ stuff so its a little less bleh.
I've been keeping within my usage because ive been in a funk a bit, but when i was slightly more worried id sorta just juggle whether claude or gpt would handle writing some initial tests as it did seem to kinda be imbalanced otherwise. seems like gpt just spam resets weekly usage throughout the week anyway so its prolly nbd.
I wouldn't be surprised if folks start complaining to California government agencies like the Department of Consumer Affairs, and they take it seriously.
There is a lot of political capital to be earned by appearing to be "tough" on AI companies.
> Claude has become practically unusable for Pro users in the past few days. The Opus 4.7 blew through an entire 5 hour limit in one question and didn’t even finish answering it
Glad I’m not the only one!
I’ve been limited so often this week I’ve setup half a dozen token compression tools in my workflow and had to do a crash course in token optimization.
Of course, it seems to only slightly delay the inevitable and doesn’t really solve the problem.
I thought it was just me. Even 4.6 is hitting limits so much sooner which I would have expected of 4.7
I hope people finally see why big tech is spending hundreds of billions in data center build out.
I have to guess that they're compute limited somewhere or the new models are incredibly overusing tokens, so I guess you need to wait for new data centers to come online?
I think that Anthropic has capacity problems. They went all in on acquiring new customers but now they don't have enough capacity to both serve users and train new models, so they are trying to limit user usages.
It is pure speculation of course, but I don't have any other explanations on the stuff they are pulling in the last 2 months.
Yeah, this is pretty clearly what's going on, but I wish they'd be more transparent about it. Funneling compute to Mythos and Design, while auto-setting effort levels lower and removing user control of extended thinking. I don't think the need to shuffle compute around is unique to Anthropic, though. I suspect it's part of why Sora got killed. And everyone's having uptime issues. Are we reaching the limits of the available compute?
So their minimum workable offer for devs just went from $17 to $100. Also, I don't see how the Pro subscription is relevant anymore. Nobody pays $17 a month just to chat. I just unsubscribed. :) Time to try Chat GPT Codex, which even works with the free subscription (don't expect crazy token allowance, of course).
> I just unsubscribed. :)
Existing subscriptions are not impacted according to Tweets from their team. It’s apparently an A/B test they’re rolling out.
If you actually wanted the $20/month Claude Code plan you may have just shot yourself in the foot.
Who would want Anthropic's business after they broke user trust?
If true, very strange change when Codex (at both 20 & 100) is a much, much better deal for a model much better at most coding tasks, with way more usage even with the /fast mode enabled. Is losing most non-enterprise customers the right move for them?
Equally, will offering a presumably unprofitably large quota of Codex tokens at $20 to retain non-enterprise customers turn out to be the right move for OpenAI?
Would not be surprised to see OpenAI follow suit.
Or perhaps OpenAI's LLMs are just so more compute efficient that they can actually offer that sustainably...
Feels to me it's a battle between who has the most compute. OpenAI does not seem to be struggling with their x2 usage on the new 100 Plan, which is very close to unlimited usage with the best performing model on the highest reasoning setting. Not mentioning the resets every 1 million customers, or the other generous usage multipliers last months. Meanwhile Anthropic seems to be desperately trying to cut down on inference with their changes to reasoning effort and more lately, so they might be focusing on what they consider to be more valuable customers for their long-term strategy. The 20 plan with Opus had gotten so bad on CC they might've just pulled the plug to stop people from complaining about usage limits. If OpenAI can burn money longer and capture the market from the bottom, I think they'd win in the long run.
Seems we were both right: https://x.com/thsottiaux/status/2046740759056162816
"Codex will continue to be available both in the FREE and PLUS ($20) plans. We have the compute and efficient models to support it."
Both the compute, and the efficient models.
Because I believe things coming out of Altman's organization at face value. /s
I mean I hope it's correct, but given his history I'm not betting on it.
What does " Not mentioning the resets every 1 million customers" mean?
>To celebrate 3 million weekly codex users, we are resetting usage limits.
>We will do this every million users up to 10 million.
>Happy building!
https://x.com/sama/status/2041658719839383945
Last reset today, after the 4 million users milestone.
> much better deal for a model much better at most coding tasks
The first assertion is also subject to change (and likely, if this works for A\). The second assertion is very much subjective. I haven't found that to be the case, but everyones needs, use cases, and workflows are different, so glad that it's working that way for you.
That's exactly what I fear- that Mythos/Glasswing has made anthropic confident that they can survive by only serving that type of customer. Would be sad to see.
It cracks me up when I hear takes like - 'if you're not using more than $20, the product isn't for you because you're not a real user.' If you use CC as an assistant rather than a replacement for your own thinking, follow SDD, and use the tool thoughtfully, you deliver a lot more and you don't need the 5x or 20x limit. It's a different story if you're vibe coding, but then we're not really talking about AI-assisted work - your three prompts barely count as doing any work. I've been on Pro for 2 years, but if this is how things are going, I'll look for an alternative. Luckily, there's plenty to choose from.
Seems like a pretty bad business move if it's really what they're doing. They should want devs using the product on a cheaper subscription to see the value with profitable limits on usage.
I think the only reason to do this would be that they just can't scale up to service the volume they have and need to cut down significantly on the total number of users. Seems also like a rough business proposition. Most of the pro plan users would probably migrate to a competitor at a similar price point (I know I will).
The only other possibility would be if they are losing too much money on the compute power and just can't offer it at that price anymore. But then upgrading the plan gives you more compute per dollar, so maybe they're just banking on people not actually using all of what they pay for?
I had previously thought that the inference cost of using a trained model was relatively low and that most costs went into training new models, but maybe that is less true with the more powerful newer models.
If it costs a ton more to serve Opus vs serving something like Kimi or Qwen, then I think most people just won't use the more expensive version for most things.
if this is accurate, and not some "oops we made a vibe-coding mistake updating our website" I am going to hit the "cancel subscription" button so hard that my desk will break in half.
I have an unlimited-usage API billing plan through my dayjob, but for obvious reasons they don't allow piggybacking personal usage onto that. so I paid for the $20/mo personal plan as an easy and relatively cheap method of professional development / keeping my skills current. I don't particularly mind paying $20/mo, but I'm absolutely not paying $100/mo.
also, part of the reason I didn't mind paying for the personal subscription is that I liked having consistency between the tools I use for my dayjob and the ones I use for side projects. if that goes away, then I might as well switch away from Claude usage at work as well. I very much doubt Anthropic's revenue predictions for this change are taking things like that into account.
making a change like this without an announcement, just sneaky updates to product pages, is also an absolutely unforgivable thing to do, in terms of me trusting them as a company.
The age of AI seems to forget some lessons from Google (and history in general).
- Rapid changes hurts the trust of your brand and product. In Google case, using a new service product became something you’ll think multiple times as you are more likely to axe it than rivals or specialized equivalents.
- While models currently has no clear winner. Anthropic’s core product is coding. But just as Skype, IE, Netscape their can always be another game changer you cannot count.
- The Pro plan is already limited for true agentinc workflows. The limits now are so bad that a business that relies on it would need bigger plans.
- Anthropic is already in a delicate situation where many devs are frustrated. Dropping or crippling the use even more just means this sector (which I can only assume is a big chunk) would switch to competitors tool that already try to compete.
- Local models, whether as Google sees it “edge” or even further would also take bigger part in the future.
“The limits now are so bad that a business that relies on it would need bigger plans.”
Isn’t this the goal to some extent? They’ll probably have the standard “light” usage plan for weekend warriors or normal folk looking to play around. Companies that mandate usage and provide the subscription for hundreds of employees will have to cough it up, and will have no problem doing so if they want to compete with the others (or so the hype would allude to).
Claude Code page still shows it included with Pro/Max. https://claude.com/product/claude-code
Claude Design was iterating on the plans page and decided to remove clutter and their review bot LGTM’d it as “minor copy change human review not required” and auto-merged it.
Shits confusing... I'm using the Claude code vscode plugin, yet my account page says I'm not using Claude code... so am I, or aren't I?
All I want is a reasonably priced subscription combining both coding AI and general AI in a single bill for non professional use that allows me to opt out of my data being used for training. Google limits history to 72 hours if you opt out of training even if you pay them $20 a month which rules them out for me. I guess I'm going to try the $20 chat gpt plan. At this point I am wondering if I need to accept that were moving to a token based model and get comfortable with opencode and manually switching models.
Chatgpt $20 plan is a steal. There's nothing close.
It's still included here
https://claude.com/product/claude-code
Maybe they're putting out a weather balloon to test sentiment. That way when they're caught they can just point at the other page to say it was just a mistake.
Confirmed, 2% of users don't see Claude code included https://x.com/i/status/2046724659039932830
As if we could just blindly trust what he said :/ Could be 2, could 30, who knows (well, they do, but not us)
Well, if there's one way to show that you're not profitable on inference, this would be it.
Either they baited people with code and flexible usage limits until march and this was planned or they realized that they did too good of an product and it costs them too much.
One thing is clear, Anthropics communications and leadership is horrible. You don't launch or remove features like this. How this is communicated and handle is something like mom+pop shop would do.
I’ve got nearly 10 months left on my yearly subscription, I wonder what that means for my access.
Same. I hope we're grandfathered in. Otherwise current pro subscribers who signed up with the understanding that they could use it in Claude Code are going to be extremely pissed and go off and sign up for alternatives (or start running local models instead). I mean, I guess they could say too bad, they got your money, but this would destroy their brand among people who are currently their most loyal users.
I don't want to leave, but I'm ready. The entire reason I got a Pro sub was so I can use Claude Code instead of going between browser and editor.
Or also sue in fact or demand refunds.
Same. I'm not a dev but I use CC a few times in a week and it's been a great help.
However, my company paid for my annual subscription, so maybe I'll ask our lawyers for advice - the only reason they paid for this was my access to CC and with my use the next tier wouldn't make sense, AND no one will expect Anthropic to not nerf it too.
Before you have lawyers look at it, wait until you’re actually impacted. Nothing has been removed from existing subscriptions yet and their employees Tweeted that existing subscriptions aren’t impacted.
Thank for the insightful comment and indeed I wouldn't without the evidence form myself and other people on the same subscription (all non-devs so we use it rarely but it feels like a superpower).
If it was removed from the pro plan, then the max plan should list claude code as one of its extras, and it doesn't.
I would not jump to conclusions yet.
In the “Compare features across plans” section, Claude Code appears to be available only on the Max plan.
You're right. I didn't scroll down. I wonder why they didn't update the top cards that everyone see. They do it for claude Cowork but not claude code? That is not very transparent. How does it make sense? It's not like claude code is too niche to be included, it's in the main app and I know multiple non-techie people who use it.
If you scroll down, you can clearly see that the Pro plan has an "x" on Claude Code now.
Previous page, when Claude Code was included:
https://web.archive.org/web/20260421141017/https://claude.co...
Edit: fixed the url thanks to scq
> The Wayback Machine has not archived that URL.
Might have been taken down?
There seems to be some JS on the page that messes with the URL. Try this one: https://web.archive.org/web/20260421141017/https://claude.co...
I assume this has to do with the $20 tier now running out of provisioned tokens so quickly as to be not particularly useful, giving users a bad experience.
The million token context + reduced caching period + new models using more tokens made this a probably unpopular but perhaps unavoidable development.
There's a hard problem here balancing costs and experience. I'm afraid despite the bad experience for people that this is necessary and $20/month was just too big a loss to sustain.
> $20/month was just too big a loss to sustain.
Is there any marginal cost associated with a new subscriber?
I have always heard inference is cheap and the cost was in training, so I assumed any subscriber was making them money, just not enough to cover their insane fixed costs.
But I am just guessing.
There are two extremes, the "approx 0 marginal cost" camp and the "anthropic losing $5000/mo on every Max subscriber" camp.
I strongly suspect both are wrong.
Next they’ll slowly reduce how much CC usage you can get out of the $100 Max plan, then introduce a new $300 “Max Plus” plan with “40x” usage. “You asked, and we listened: Introducing Max Plus, our biggest plan yet, designed for those…” blah blah
I can't believe they are yanking tool access instead of just reducing the token quota or simply pulling Opus 4.7 access. To be fair even that would be poorly received, but at least people would have a choice of working within limits. Claude Code is their real winner, and a great ramp for newcomers coming into AI assisted development. They are playing straight into OpenAIs hands.
Anthropic NEED to get better at communicating with their customers. The most meaningful updates we get on changes come from employees on X. It's unprofessional and unsustainable.
Pricing page reflects this: https://claude.com/pricing#individual-features
That is exactly what TFA shows
Only for some, though. My pricing page still shows Code on the Pro tier.
Just tried Claude Code on my Pro plan. It worked. So no, it's not removed
I would assume users who have an existing subscription will be grandfathered in.
It would seem misleading to sell monthly, or even yearly, subscriptions under the guise Claude Code comes with the subscription, for it to only be yanked out underneath you. (Although depending who you ask, Anthropic have already done actions similar to this).
I was billed $220 on Mar 1 for the pro plan
If they rugpull Claude code from my already paid for annual subscription I’ll have to issue a chargeback
The $20 plan showed me how good Claude Code was, and now I'm paying $100/month. I never would have paid $100 just to try out Claude Code.
Many, including me tried the $100 just to try Opus (wasn’t available of pro before) and first thing I thought was “worth it”
Let’s say my trust level and appreciation for the product for the past month had a big negative hit for me
Anthropic clearly doesn't understand that customers see their brand as "Claude", Google's brand as "Gemini" and OpenAI's brand as "ChatGPT." They have so many plans and exclusions that they risk customer confusion. I was surprised when I was pay $200/month for Claude Code, finding it super helpful, and then I had to pay separately to get API access for an experiment. Why are so many parts of "Claude" separate from each other, especially on a $200/month subscription.
Anthropic better get this sorted out with a proper product manager and marketing or they risk customers jumping to easier to understand platforms that are good enough.
>or they risk customers jumping to easier to understand platforms that are good enough.
https://www.anthropic.com/news/anthropic-raises-30-billion-s...
I reckon they'll be fine. Not agreeing or disagreeing with you, but they have enough customers who won't leave.
That also was really opaque to me RE: API access. I initially thought at $200/month I could get whatever I needed. I eventually set up a OpenAI API with a few bucks to try what I wanted to.
https://claude.com/product/claude-code - scroll to "Get started with Claude Code" and it's still lsited in the pro plan.
Ed’s replies show an archived Code support page that was changed today to remove mentions of Pro. They seem to be making these changes right now.
> making these changes right now
Would it really be that hard for them to just make all of the changes and then do a redeploy rather than doing them incrementally? It's not like they're just editing the raw HTML sitting on the server manually, right? Actually, don't answer that, I'm not sure I even want to know the answer.
“Claude, remove the reference to Code in the pro plan everywhere on our website.” “Done” “You forgot this page.” “Done” “You also forgot this one”
3 hours later…
Hey, I'm a pro, and I feel genuinely insulted. I could consider going back to Claude Desktop + MCP, but I'm getting tired of this telenovela, and will probably cancel my sub and take my business elsewhere.
With GitHub and Anthropic reducing subscription features, Chinese providers are looking more and more tempting.
Until you work for a company or government agency that is subject to any sort of technology audit. The moment offshore processes running in China comes up you'll have a never ending hole of questions to answer.
This just made me gamble on a yearly subscription for Pro, hoping they will grandfather in existing customers..
Please report back if they send you a Dear John email.
Hmm, we just bought my wife an annual subscription at the Pro tier, largely to use Claude Code. Wonder if she'd be grandfathered in or if we'll need to get a refund.
Sooner or later, Anthropic will probably have to take this step if they want to actually make money. Either way, they're going to lose a huge chunk of their user base because of it. I currently have a private Pro subscription for $20; I can just barely justify that amount for my personal hobby projects. Work-wise, my employer provides all of that for me, so I don't really care - this is just for my personal needs. But at $100, I'm out. The subscription isn't worth it to me just to run a few prompts in Claude Code once a week (yes, I'm still one of those people who is actually capable of, and enjoys, writing code by hand). I do use Claude Chat as well, but the free plan is more than sufficient for my needs (never ran into any limit until now). I simply have zero interest in Cowork and any products other than Claude Code.
In short: if Claude Code is removed from the Pro plan, Claude in general will no longer be interesting for people like me.
I hope they really botch Claude Code to give all the developers a wakeup call about making a hard dependency to a for-profit entity that's essentially in it's startup phase. Like, we wouldn't adopt some newfangled framework just because it was on the front page of Hacker News. Why did we all jump on the new unproven technology here?
I predict this may get reversed as it would be a huge opening for glm, kimi, and qwen offerings. I'd switch instead of upgrading to Max
Do they have their own cli agents or just the api inference services?
I don't think they do but you can always use OpenCode or Pi Agent.
very easy to configure claude code to route to GLM as well.
Moonshot AI has Kimi Code and their $20 per month plan had a lot more than Anthropic's (before 4.6 and the other changes to eat up your limit faster).
Some do, I know of kimi code https://www.kimi.com/code
Time to do research.
This is a joke, right... right??
Trying to chat with "fin" is like trying to argue with a bowl of congealed oatmeal.
Me: Why is your website saying Claude Code is not included in Claude Pro subscriptions?
"Fin", the slow, dopey AI agent: That’s confusing, but it usually comes down to what “Claude Code” means in that context. Claude Pro does include Claude Code (the Code tab in the desktop app, and Claude Code access for individuals).
What Pro does not include is API access via the Claude Console. Those are separate products: Pro is for chat, and the Console is for building apps with the API.
So if the website is saying “not included,” it’s likely referring to Console/API access, not the Claude Code feature itself.
Did that clear up the confusion?
Me: No. It clearly states on the pricing page that Claude Code is not included.
Fin: That’s frustrating, <NAME>. The key detail is that Claude Code is included in Claude Pro, but API access via the Claude Console is not.
So if the page you’re looking at says “Claude Code is not included,” it’s almost certainly referring to Console/API access (the developer platform), not the Claude Code feature itself.
If you tell me the exact URL or screenshot of the page that says that, I can confirm which “Code” it’s talking about.
Was that explanation helpful?
Me: https://claude.com/pricing
Fin: On https://claude.com/pricing, Pro does include Claude Code. It’s listed right there under Pro’s “Everything in Free, plus” bullets: “Includes Claude Code.” So if you’re seeing “not included” somewhere else on that page, it’s likely referring to something different than the Claude Code feature. The one thing Pro does not include is API access via the Claude Console (that’s a separate product).
Also, the Code tab in the desktop app requires a Pro (or Max/Team/Enterprise) subscription, and it may not show for everyone yet since it’s still rolling out.
Does that clear things up for you?
Me: there is clearly an "x" at this page. escalate to a human immediately.
and then the cycle continues. I'm cancelling my Anthropic subscription, sick of worrying about limits and getting banned and simply awful experience.
[dead]
At least for me Claude Code is still working on my Pro plan. I don't know if that's because the change simply hasn't propagated all the way through their systems yet (the change is now up on the main Claude pricing page and on their support pages, but not on the Claude Code landing page yet), or if it's because existing plans are grandfathered in, or what.
In general Anthropic seems to be pretty bad at clearly communicating what is going on. I have both Claude Pro for Claude Code and ChatGPT Plus for Codex, and lately I've been reaching for Codex first more and more often... at least for the hobby stuff I'm using Claude/Codex on, they seem pretty much equivalent in terms of practical capability/usefulness.
How long until OpenAI remove Codex from their cheap plan?
Should we instead use a generic coding agent with a particular model and just pay per token?
pretty sure the codex cli itself is open source (and written in rust!) and can be used with any model.
This is more ethical than what they've been doing, trying to keep those subscribers but limiting them to the point it's become unusable. But it's also kneecaping themselves because they'll miss out on any innovation and hype coming out of the hobbyist community.
Simultaneous with GitHub copilot dropping support for the Opus models in their $10/month plan...
I'm curious about their expectations and how they will interpret the results.
On the one hand, the people there are supposedly among the smartest on the planet. On the other hand, they consistently forget that they're dealing with LOYAL humans, and these humans prefer respectful communication beforehand instead of being messed with every other day.
My hope for reasonable behavior is to not handle it this way. Decrease limits and increase prices if you can't handle it and be _honest_ about it.
Are they just looking for a way to rationalize another hostile act? And already have expectations like:
- "minus 10% in pro signups" -> oh, let's drop those coders who won't pay anyway
- "minus X% in pro signups and plus X% in max" -> awesome, PAY UP!
https://www.wheresyoured.at/news-anthropic-removes-pro-cc/ wrote it up. And relatively dry too!
The rug pull is coming
Max is next. He essentially admits to it in one of his tweets/posts. Explicitly citing it as an example of how they misjudged usage relative to pricing.
I wish they actually start doing $500 or $1000 a month plan to avoid having to use so many accounts to get work done.
ANthropic never wanted my money anyway... they don't allow work + personal accounts to have the same phone number. I had to close my personal account otherwise I could not complete onboarding at work.
So I pay for Codex instead.
You need to onboard with your personal phone number at work?
Why not with email?
Anthropic forces everyone to verify both.
I have a seat in my company's enterprise plan and never gave a phone number.
Maybe it depends on the plan. We are on Team not Enterprise.
You should be blaming your employer for forcing you to use a personal device to access company resources. You should have been given a company phone or stipend.
I did complain about it in our chat and my boss later reached out, offering to set up a number on our company's Twilio account but by that time I had already entered my personal number.
UPDATE: it appears this is no longer a problem. Just tried again today, after verifying the number they ask if this is a personal or work account, I selected "Personal" and now I can use Claude both for work and for my own stuff.
You can just get numbers online, no need for SIMs anymore nowadays.
It is perhaps better for Anthropic to do a price hike e.g. $25 or $30 for Pro with a clear/honest messaging e.g. "Running costs are high, price hike is unavoidable" than resorting to these tactics.
These shenanigans are earning them no respect. The market is already annoyed on model serving QA issues, and now (recently) Opus limits. They don't want to lose to OpenAI - understandable - but these shortcuts won't earn them anything either.
The last couple of weeks using Claude has been…interesting to say the least.
Additionally I run a constant hacking contest between GPT and Claude. It’s a toy project and it simulates an attack/defense of a small corporate network.
Claude used to win pretty handily. Suddenly it’s started to lose 90% of the time. I thought GPT had gotten better but no, looking at the logs it seems that Claude is slower and more prone to running in circles. This is still the case when switching to Opus 4.7.
I don’t know what that means but it’s undoubtedly worse.
https://claude.com/product/claude-code seems to say it is included in Pro plan.
They should allow existing Pro plan users to keep using Claude Code.
Everyone that is upset about this should take note: you are not a (coding) customer at $20/mo. Their coding customers spend thousands per month (week!) on claude and it's growing faster than they can keep up with (source: I'm one of them, and I know many other like me. We're budgeting 10-20% of engineering salary spend on tokens). It sucks to no longer be able to code on the cheap anymore, but don't fool yourself into thinking you have any leverage here.
I agree about the leverage point, but it was definitely possible to be a coding customer at $20/mo before, especially with Sonnet.
My leverage is i’m canceling my plan. Openai gives me codex+chatgpt for my $20. I use my claude code sparingly, but I enjoyed it and it works great when I needed it. Access to it was a reason they got my money a few months ago and it’s been a shit show of reduction of services since.
Note that some companies, like Amazon, purchased and ran the Claude on their own hardware. They didn't change the model parameters during the Claude Opus 4.6 karma.
If Anthropic continues to getting worse, try Amazon Kiro and other companies that run Claude on their own hardware.
It might be expensive and have a worse experience compared to Claude's code, but at least the model itself is the "original flavor."
These days, it's hard to ask for much.
I know people who work on Kiro, happy to take feedback. It'll get better.
I don't see it mentioned in the Max plan there either.
Feels like a very strong chance that this is related to trying to find a way to navigate letting OpenClaw users on. https://docs.openclaw.ai/providers/anthropic https://news.ycombinator.com/item?id=47844269
Anthropic should be increasing the value and services they offer not reducing it.
It’s the Apple model. Yes you pay a ton more. But my 2013 MacBook Pro 15 I got in college lasted 10 years and was still fine even when it was stolen. That’s what you pay for. You pay for a ton of built in apps and functionality and quality.
Arbitrarily removing things is customer and more importantantly good will hostile.
Oh FFS claude code is the only reason I have a pro claude subscription. I don't even use my personal subscription all that much after spending all day with claude/bedrock at work. I will absolutely cancel my pro subscription and continue to use local / Codex if claude code stops working.
I realize this duplicates a lot of sentiment already in this thread but anyone here with pull at Anthropic please understand it will undo a lot of the goodwill that made Claude so successful in the first place.
Maybe you can try Kiro. Happy to take feedback.
And this is why local AI is going to be the future.
Local AI is almost impossible right now with the prices of RAM and GPUs and the sizes of decent models. No way spending even an optimistic 10k, but more likely 20k, on a setup that is good for 5-6 months makes any financial sense.
Not yet, but it's only a matter of time. Local AI doesn't need to be bleeding edge either. Even capabilities of older modals would be fine.
I disagree wholeheartedly. Older models do not perform anywhere near as well as newer models, and certainly not once you throw in agents that can sense check, security check, refa tor and balance, and research queries as they run behind the scenes.
Just get a Mac?
Impossible to find Mac minis in some areas, and if this goes through expect it to get worse.
I settled for the AMD rough equivalent. It’s not perfect but it can still handle most of the work. Now if only extra ram would come down in price… I find I need about 5 GB more than I have
Macs will just get more expensive. C'mon man, we're supposed to be Silicon Valley engineers, the pinnacle of human intellect.
It’s seems like there are a lot fishy smells coming from the timing of the mythos announcement and the reports of issues with casual users. Combine that with the mass rejection of 4.7 it kinda seems like they are burning their ‘non research’ users in order to keep the Mythos users warm.
I could be connecting unrelated dots here, but it sure as hell seems quite coincidental to me.
To help you decide if you should keep your Claude subscription, you can see how much of your code is written by Claude Code with my project (open source, local): https://github.com/gelatinousdevelopment/buildermark
The clusterfuck about the various pages not being all upgraded to reflect the same story...
I thought we now had advanced tools to which we could ask to do things like: "Remove all mention of Claude Code in the Pro, but not in the Pro Max plan".
But apparently the CGI-days called and asked the webmaster to manually edit .html files one by one?
From Amol, who is the Head of Growth:
> For clarity, we're running a small test on ~2% of new prosumer signups. Existing Pro and Max subscribers aren't affected.
https://x.com/TheAmolAvasare/status/2046724659039932830
But the current plans are unsustainable and prices will have to be effectively raised sooner or later:
> Engagement per subscriber is way up. We've made small adjustments along the way (weekly caps, tighter limits at peak), but usage has changed a lot and our current plans weren't built for this.
https://xcancel.com/TheAmolAvasare/status/204672528250217304...
This is the dumbest PR tactic in the book, and it annoys me that it works on so many people.
April: "The fact that we're doing X isn't news because we're only starting to do X"
August: "The fact that we've fully rolled out X isn't news because we started X in April"
Once again random tweets from insiders being the only clues we have to what Anthropic actual policy is
you could try customer support, that chat bot will happily loop you with some more non answers, but try to make you feel good about those non answers :)
This makes sense given Anthropic’s recent downtime and resource constraints.
Opus 4.7 consumes tokens at a faster rate and folks were complaining that the Pro plan included too few credits for real work.
And Anthropic now allows `claude -p` (which invokes Claude code) for 3rd party agents like OpenClaw, which consume far more tokens by running autonomously, 24/7.
only if it actually improves the downtime, people were expecting the same when they revoked openclaw access but that didn't change anything
The only thing they'd need to do to enjoy the positive PR from the DoD spat is shut up and improve (or at least not worsen) product.
Even the downtime would've been fine (as GitHub shows). Instead they're pissing it all away by letting employees make random announcements on random platforms.
This is exactly what happens with all agentic products in near future. Demand is way higher than supply right now. Also most programmers do not use agents yet so demand will grow even more. Building data centers is not easy and price of HW is skyrocketing.
I see lots of speculation that Anthropic needs to cut usage because they are compute constrained. If that's the case, will they be focusing on reducing compute costs for their models?
From what I can tell Opus 4.7 is more resource-intensive than Opus 4.6 is more resource-intensive than Opus 4.5.
It works for me at the moment on the pro plan. Is there a grace period until they enforce the new pricing?
I would hope that we'd be grandfathered in since we signed up with the expectation that we could use it in Claude Code. I could see where maybe that might lead to problems down the line where they do some kind of update and "forget" that people who signed up before a certain date were supposed to be grandfathered in.
Are compute resources so tight they’d want to make their products as unappealing as possible on purpose?
Related perhaps to GitHub pausing signups for Copilot? https://github.blog/changelog/2026-04-20-changes-to-github-c...
I was seriously considering investing in an RTX 6000 rig, and I think this is just enough to tip me over the edge. I know it's not as good as CC, but at least I know it'll be there tomorrow.
Vibe coded ui bug?
That sucks, I guess I'll cancel my Claude account. Not paying 100 dollars. That's crazy
Is Claude Code actually useful on the Pro plan?
I remember when they first added Claude Code to Pro — it was limited to Max initially — and my first thought was that it seemed kind of stupid, because at one fifth of my current limit, I would be hitting walls all the time...
I would say it's definitely useful for side projects. You have to manage your context windows a bit more but it's still definitely worth the money
I’ve found that I hit the limit just around the end of the 5-hour window, so it’s definitely been usable for me.
But I’ve mostly been using it for gitops infrastructure in my homelab. I wonder if the token usage is lighter than if I were developing an application.
It was for about the first 6 months after I subscribed, then the rate limits were tightened to the point of uselessness and pushed me to cancel and go for the Codex plan instead.
I use it on Pro and was just thinking today, there is no way $20 covers the cost of it. But I'm long term unemployed and can't afford any higher tier, so if they drop it guess I'll have to find a non-anthropic solution somehow.
Sonnet in the Claude Code harness is hard to replicate out of the box. That vertical integration is not easily replaceable but by no means impossible...
OpenCode and their Go plan will get you close if you're willing to put in the config work.
For when you do need the larger models Fireworks has a pretty generous 'Pass' that comes out to about $7 a week for some of the larger bleeding edge models.
Other than that Codex's $20 plan is still somewhat valuable though they keep reducing usage. Google's $20 plan will get you some Opus usage in Antigravity and a generous amount of Gemini. Not sure how long that will last as they've been tweaking pricing and planning language recently too.
What do you mean "somehow"? There are plenty of alternatives in the market.
People really have to internalize that these things are expensive to run, and that there isn't enough compute to go around, like actually for real for real, which is likely the source of this. My guess is its a temporary new sign up pause.
Unrelated to the Claudge Code change, I'm fascinated by people on Twitter and Bluesky posting screenshots of the answers they get from AI like it's an original source of information. It's as if some users see the AI as an authority, and derive some kind of social capital from that authority. For example, in the OP's linked Bluesky thread, one person replies with "Fin says it’s included with Pro" and attached a screenshot from "Fin AI Agent" (which I haven't heard of) that claims Claude Code is still available on the Pro tier. Is that valuable? Personally I don't trust what any AI has to say, especially when the subject is currently in flux.
https://bsky.app/profile/mattgreenrocks.bsky.social/post/3mk...
Another example, I recently saw two people over on Twitter posting LLM responses at each other in a bitter argument about Vercel's security breach. They made no attempt to pretend they'd formulated the ripostes themselves, it was just screenshotting one-sided conversations... What's the point? They could've saved themselves the trouble by spawning two LLMs, naming them "John Doe" and "Fred Doe", then telling them to argue and post the name of the winner.
Disclaimer: I don't use Twitter, Bluesky, Mastodon, etc., so maybe it's not that deep.
Can't you use Claude Code without a Claude model? E.g. can't you use this with a local model?
Folks are assuming that only the $100 plan will include Claude Code access. I think a more likely scenario is that everyone will be able to use CC.
I'm very happy with GLM 5.1, MiniMax M2.7 and Kimi K2.5. My wallet is happy, too.
[dead]
The "5x" and "20x" no longer make sense for Max. It's supposed to be 5 times the Pro limits. But if only Max 5x has access, then they need to renamed to "Max 1x" and "Max 4x".
5x 0 = 20x 0 = 0
Maybe this is coming next
"We've determined that claude code is too dangerous to your code base to release, so we are withdrawing it"
It could still mean 5x the Cowork limits.
They'll remove it.
Then some genius intern will say "if we offer it to the lowest level plans, the users will get hooked on how awesome it is!"
Then Anthropic will put it back.
Or they could just ask Claude if it is a good idea to remove.
How long until the $10 Github Copilot subscription goes away? That was a great deal for my limited personal programming. The only reason I switched from it to Claude was to get coding and general ai in a single bill.
https://github.blog/news-insights/company-news/changes-to-gi...
I think Github Copilot is in the process of slowly winding down right now. They've been putting very, very long (multiple day) rate limits on users for various esoteric reasons for weeks now and just yesterday or so paused signups.
It already did, you cannot sign up for it right now.
I know this would add to pricing/model/plan confusion but what about Claude Pro plan can only access Sonnet and Haiku Models via Claude Code?
Opus is fairly useless on Pro given the rate limits anyways.
And of course you never want the first tier it's included on because you need actual usage, so in reality you need Max 20x - the price of which I can't even see on the pricing page
Tangent, all those people who are coding in extreme niches, like embedded development, is AI helpful? Can it learn enough from stealing FOSS code to help?
so if you pick max-5x or mx-20x than pro... are you getting 0 now?
I was working in my biggest project in the last 3 years and i dont got mutch money and they go there and cut claude code from pro like wtf so why am i even paying for pro
Claude Code still works on Pro plans as of right now.
The Anthropic website has become inconsistent. Some places say Claude Code is included in the Pro plan, other pages don't.
New sign-ups only or affecting current subscriptions too?
Are they trying to reduce cloud servers needed for AI because they're running out? I wish they'd be a little more honest in their response.
Cool, first remove OpenCode support so I have to use their shitty TUI, then remove that too. What a loss. Good thing GLM 5.1 is good enough
This makes me less likely to choose Claude Code, as this feels unreliable... Who says they will not change a few months down the line again.
Practically speaking, the $20 plan is useless anyway given the limits and the way Claude Code consumes tokens with very little caching.
We, as a society, need Open Source SOTA models and should be able to also train new models via distributed P2P network.
Otherwise companies will keep exploiting using their dominant position.
We were taught dictators are bad, monopolies are bad, but now allowing 2 companies control most of the software development
That seems not possible.
at the moment - yes.
but why not work towards it?
* elect politicians who will support this
* change laws to accommodate it - if corporations train data, on every usage they should pay higher taxes so they can't exploit the open data, but public ledger trained model is fine to use open data
* similar tech exists (bitcoin, torrent), needs some modifications
I'm out of the loop. Do you guys use Claude subscriptions for Claude Code rather than just pay-as-you-go API keys? Is it cost related?
Yes, it's been a way better deal to go for a subscription than pay as you go for me in the past. I had a month where I burnt through ~3.8b tokens which was somewhere in the ballpark of $8k worth of savings.
Now though I don't dare use spend tokens for basic note taking with Sonnet because I'm hitting the limit over a couple million tokens on the 20x plan, so they've really tightened the purse strings since November.
I mean one is a flat 20 a month and pay as you go is going to go over that for any heavy coding use, right?
Dup: https://news.ycombinator.com/item?id=47854477
They will lose the individual subscribers for sure. My day job can pay up to the roof for ai access in this climate but personally I cannot.
Believe me, plenty of "day jobs" are already stretching to pay for $20/mo for their users.
Others in non-tech sectors are forced to use Copilot. Who knows what I would pay for a usable LLM out of my own pocket. Probably more than $200.
Will definitely discontinue my Pro subscription if it's true. Claude Code is the only reason I signed up in the first place.
Every person who signs up for a pro subscription on that advertising has cause to sue for being miss-sold the product.
Frontier LLMs are not going to be economically viable when the free money runs out and they're expected to turn a profit.
Claude Code has become useless anyway in the last few weeks. I want my money back!
Curious how this will play out (if true) for folks who signed up for that annual plan - expecting Claude Code to be included.
That’s too bad, I just purchased a one year pro plan for my dad just for him to play around with CC when he retires next week.
I'm not going on max. I barely hit 20% of my weekly limits on pro, to hell with Anthropic if they drop this plan.
So this is why they allowed Claude code to be used for open claw again - cuz code is going to more expensive customers
Does this mean that for enterprises using per-seat pricing, only the $100 premium seat gets access to claude code?
Team plan shows “Claude code” in a main bullet point still. Which would indicate it is part of the team plan regardless if it has premium seats or not.
But it seems this is all in a state of flux.
And there’s the lovely asterisk at the bottom:
> Prices and plans are subject to change at Anthropic's discretion.
Enterprise doesn’t have premium. Just api usage.
Business accounts are like max 6x accounts.
It's going to be interesting seeing the AI companies gouge people who are incapable without it.
How about team plan standard customer?($25) I suppose its only natural it would follow suit if not already
They have till morning to clean up this giant PR mess. I think people will leave for codex
I can't even sign up for Max (last tried yesterday), their credit card processor has issues.
I just bought my brother 6 months of pro so he could use Claude code. How do I request a refund
Chickens have come home to roost. Someone had to pay for the servers and the debt, and its us!!
Run your LLM locally.
That is the only way to avoid being held captive by Anthropic / Meta / Google.
Claude Code started to suck since a couple of weeks ago. I want my money back!
Bubble stress test, stage 1.
i cancelled cursor about 10 minutes ago and i'll cancel claude too
Back to coding by hand like our ancestors? Good for you! We should all do this, give them the finger.
Or you know, you can use a local model, or an open model with an independent provider.
Or, coding by hand :)
Funny, I just signed up for Pro a couple hours ago, to check how Claude Code works using this plan, instead of using my API keys.
I got rate limited after about 30mins of coding and was thinking, who the hell i going to work like this?
So they really seem to be running into extreme capacity issued now.
Anthropic reverts the “prosumer” AB testing, but damage is done.
Their chat bot claims that pro users still have access to Claude Code.
"You are absolutely right!"
I just cancelled my plan, but still have access to Pro and Code apparently until my cycle would have renewed. Hopefully they get a clear signal from this, especially if more of us cancel with the intention to sign back up should they reverse this decision.
I’d be surprised if they’re running at less than 100% capacity after this. It’s just too useful to too many people for whom an $80/month increase is immaterial (I speculate)
This is not so much about capacity, I pay $X and I get a certain amount.
This is much more about removing access to a feature I have access to. Having a chat with Claude Code (removed) seems to me to be likely less usage than Claude Cowork Tasks (more like open claw, cron triggers).
This seems much more like a SaaS Tax. I would not be as unhappy if they kept the Code feature and gated my access to Opus. I mostly use Sonnet anyway
See also https://www.reddit.com/r/ClaudeAI/comments/1srzhd7/psa_claud...
No particular opinion on this change, but generally pricing is a great way to separate dabblers from serious users. There isn’t a great deal of value in dabblers or what they produce, I imagine that training data isn’t worth much relative to the pro users. Similar pricing story with $100 yearly price for Apple developer accounts that people complain a lot about. The reality is if you’re serious about making something, these costs are pretty cheap.
The folks hurt most by this are serious people in developing countries and young people starting out. Occasionally a dabbler turns into a serious user but I imagine that’s far less likely than people wish it were.
The value to companies who make these changes is they don’t have low value users or low value contributions to worry about, which has its own not insignificant overhead. In the age of AI slop everywhere we’re likely to see a lot more attempts to separate the wheat from the chaff.
The dabbler/serious user distinction isn't the only framing here.
Assuming this limitation applies to team seats in the same way, at $20/mo, businesses could afford to have everybody on the plan. Plenty of folks write only a few hours of code per day—or even per week in their job. These are still professionals, not dabblers.
Ahh couldn’t they vibe a solution to make CC cheaper?
Is this confirmed?!
Anthropic is vibe running the company.
I'm asking for a full refund of my plan via amex if this actually happens. Fuck them.
does anyone have recommendations on replacing CC with something else for around $20-30 / month?
;) - use the Chinese models ;,>
This just lends more fuel to AI skeptics that this entire thing is a massive, unsustainable grift. The explanation only adds confusion and implicitly means that this was not a mistake. What is someone to take away from this?
That $20/month is not profitable? That Anthropic thinks that people are willing to pay a 400% markup without batting an eye? That Anthropic is desperately trying to clean up their burn rate? Why should we trust a company that can screw up basic PR this hard?
There’s also a very real possibility that the skeptics are right.
Jesus Christ, it’s literally going to become too big of a risk to depend on Anthropic.
As someone who tries to manage usage for a small team they just added Claude Code to the Standard Team seat now they are removing it!?
Not to mention that they will ban your entire organization from a bot deciding you violated their TOS with no communication and no way to contact anyone to understand what happened.
If this is real we are switching to OpenAI or Gemini it is not worth all this non sense
Annoying, signed up for a personal pro account a couple of weeks ago. No way I’m going to spend $200 just for a few personal projects.
For $20, it seems like a no-brainer to switch to Codex.
However, if you think you'll need the extra capacity of Max, it's $100, not $200.
no, most likely you will try codex and get more value (as I did)
I feel like Anthropic has managed to burn good will faster than anyone I’ve seen since Elon Musk.
Huh? I just don't understand why they're doing this. Feels like shooting themselves in the foot, given that Claude's individual subscribers are a large part of who is introducing all their enterprise customers to them. Plus removing access is never good for public perception.
You know you can just code by typing and thinking, right? Like you don’t need to use AI if you’re a software engineer.
I know, crazy idea. When we told you they’re getting you hooked and would rug pull you called us permanent underclass or something.
No way
They're trying to find every way to enshittify their partially unprofitable service. When they find a way that sticks, they'll go with it. This has become the preferred way of doing tech business in the US. Create a great thing, give it away for free, hook users in, try to squeeze them. In theory competition should limit this kind of behaviour, but for some reason they big companies all wait on another to start enshittification in unison. How this is legal still puzzles me but evidently that's how it goes.
Here's my hot take: Anthropic et al. are trying to make developing a subscription-only job, and they've done that by illegally pirating pretty much the whole Internet. If they were to go out of business tomorrow and serving models was to become a commoditized service like storage we'd be all better off. Sure, we would have less research on frontier models, but we don't need AGI, we need good local models, RAM and good open source / weight AI tools.
I was told Claude code was profitable on all subscriptions. Lol, turns out maybe you can be more profitable.
Some top tier PR yet again... does Claude run their PR team on its own?
“Let’s see how they react, and they will be ok and we will be rich.”
Wait. What is Anthropic without Claude? Serious question.
Meh, $20 Codex is better at this moment anyway.
Nice. Bye.
The enshitification stage came quick.
The enshitification intensifies.
[dead]
[dead]
[dead]
[flagged]
I posted this question two weeks ago: "What is your plan when the AI you have implemented throughout your company changes the results you've come to trust?" (https://www.theregister.com/2026/04/06/anthropic_claude_code...).
Since then, I had to add:
"or won't let you log in?": https://github.com/anthropics/claude-code/issues/44257
"or makes stuff up?": https://dwyer.co.za/static/claude-mixes-up-who-said-what-and...
"or when it's down?": https://status.claude.com/incidents/6jd2m42f8mld
"or when you get banned?": https://bannedbyanthropic.com/
"or installs spyware?": https://www.thatprivacyguy.com/blog/anthropic-spyware/
And this is all exclusively about Anthropic. It's insane. On any other tech, there would be a consensus to wait until it's stable, but not AI - we go full throttle when it's AI.
Genuinely curious how people who have implemented this in serious companies are answering these questions, because my answer is to keep it the fuck out.
Saw this coming eventually. $20/month for autonomous agents running 24/7 was clearly not sustainable at API pricing. The part that's surprising is there's still no official announcement - just a quiet page edit.
The $20/mo plan never supported 24/7 autonomous agents. With Opus 4.5 and 4.6 I would hit resource limits after a reasonable amount of work, which corresponded to a variable amount of wall clock time.
This makes me think either they’re severely resource constrained and need to focus on “high value” customers, they’re bleeding money on inference, or their sales and marketing team is incompetent.
Regardless, this feels like a pretty big rug pull. Especially without a phase-out period and a real announcement. As someone using Claude Code on a personal hobby project to get a better feel for its capabilities, I’m not sure what to do now. I can’t justify the $100+/mo plans for a hobby project.
My choices are then:
Fun times.> running 24/7
This was never the case though. There's a per week and per 5 hour quota. If you exhaust either you have to wait for the reset. What they're doing makes no sense.
Choosing to do it quietly instead of letting everyone know is actually not that surprising.
And yet they're very aware that Hacker News, etc exists and so the awareness and backlash would be instant. It's like they want to get a lower rating from the community. Maybe that's their solution for the resource issue: make enough people mad so they abandon their subscriptions.
Wonder where this leaves folks who paid the annual rate? Here’s what Claude said:
https://claude.ai/share/1a4293bd-b2d4-41b7-a887-eb42b3ae8b6e
“ The standard answer here is no — Anthropic does not typically refund the unused portion of annual plans , and annual subscribers won’t see prorated refunds, retaining access for the full remaining period instead. That said, your situation is a bit different — you’re not just canceling, you’re canceling because a feature you paid for was removed. That’s worth contacting Anthropic support directly about. Their support team can check your refund eligibility , and this kind of material change to the plan is exactly the case where a support escalation could go differently than a standard cancellation. You can reach them through the in-app support messenger at support.claude.com or via the thumbs-down feedback button. I’d recommend explaining specifically that Claude Code was a factor in your annual plan purchase. ”
Just do a chargeback its an easy W. I dont see why you'd want to continue doing business with anthropic after a change to a 1 year contract out of cycle