It's fine, their moat is thin. Frontier models as a service isn't really in the best interest of anyone anyways. Only a matter of time.

Are you arguing that eventually a competitor will emerge that does support OpenClaw with a subscription model? Wouldn’t that just be more expensive for the exact same reason Anthropic is banning it?

OpenAI have literally gone out of their way to explicitly support this sort of thing. As they did with OpenCode.

Honestly, this just looks like what Dylan of SemiAnalysis suggested on Dwarkesh – that they've massively under-provisioned capacity / under-spent on infrastructure.

That would honestly be a comforting answer if true, because I would gladly take 'we can't afford to do this right now' over 'we are self-preferencing, and the FTC should really take a look at us, even if we're technically not a monopoly right now, since we're the only strongly-instruction-following model in town and we clearly know it'.

> we are self-preferencing, and the FTC should really take a look at us, even if we're technically not a monopoly right now

Tell me you have zero clue what a monopoly is or what the law is, without telling me.

Monopoly law relies on broad categories, not narrow ones. You can’t call Microsoft a monopoly because they are the only company that makes Windows. You can’t call Amazon a monopoly because they are the only company that makes AmazonBasics. You can’t call Anthropic a monopoly because their product is 20% better for your use case, otherwise by definition no company has any incentive to do a good job at anything.

Somehow this was coming up a few years ago where people kept saying that Apple could face antitrust because they were the only company who made iOS and controlled the App Store. Given that android exists, and has roughly equal market share, that didn’t make much sense to me, but I kept seeing it being discussed.

And Apple did lose that case now so they were correct; sometimes, one can be a monopolist in the market they created.

Well, Apple did recently lose as they're the monopolist in their walled garden for app distribution.

> Tell me you have zero clue what a monopoly is or what the law is, without telling me.

Monopoly law is subject to reinterpretation over time and anybody who has studied the history of it knows this. The only people argue for "strict" interpretations of current monopoly law are those who currently benefit from the status quo.

> Monopoly law relies on broad categories, not narrow ones.

And this is currently a gigantic problem. Because of relying on broad categories to define "monopoly", every single supply chain has been allowed to collapse into a small handful of suppliers who have no downstream capacity thanks to Always Late Inventory(tm). This prevents businesses from mounting effective competition since their upstream suppliers have no ability to support such activities thanks to over-optimization.

To be effective on the modern incarnation of businesses, monopoly law needs to bust every single consolidated narrow vertical over and over and over until they have enough downstream capacity to support competition again.

Oh, give me a break. I know the law around this incredibly well. Reasonable people can disagree about whether the law is appropriate. The whole point of laws is that they should match intent – and as for '20%': "tell me you don't understand how a small quantitative gap can result in a step change in capability."

> Oh, give me a break. I know the law around this incredibly well.

Then don’t make BS up like implying Anthropic is a monopolist for the crime of competence.

> tell me you don't understand how a small quantitative gap can result in a step change in capability

The law does not give a darn about this. Being a good competitive option does not make you a league of your own. If I invent a new flavor of shake, the Emerald Slide, am I a monopolist in shakes because I’m the only one selling Emerald Slides? If you go and then start a local business reselling shakes and I’m your only supplier, am I a monopolist then? Absolutely not.

You do realize that I called out in my post they are absolutely not a monopoly by the law, right? I know all-too-well what the definition is.

We have a similar situation in mobile where Apple may not be considered a monopoly, but people have walked around for a decade with a supercomputer in their pocket that is wildly underused.

Things have gotten faster; things are different than they were decades ago when a lot of this was devised.

The reality of the matter is that some of us just want to see innovation actually happen apace, and not see 5, 10, or 30 years of slowdown while we litigate whether or not such a company is holding all the cards, while everyone is collectively waiting at the spigot for a company to get its shit together because we're not allowed to fix the situation.

For what it's worth, I'm hopeful that the other model providers will catch up and put us in a situation where this conversation is irrelevant.

What I'm afraid of is a situation where we see continued divergence, and we end up with another Apple situation.

You’re welcome to start OpenSpigot yourself, and see how investors feel about you giving away your technical / IP / market advantage on launch day.

> “we are self-preferencing, and the FTC should really take a look at us, even if we're technically not a monopoly right now”

That is not calling out that they are “absolutely not a monopoly by the law” in any way, shape, or form. You’re framing it as though they aren’t by a technicality, when they aren’t anywhere near discussion by even the most extreme of legal theories. You won’t find Lina Khan or Margarethe Vestager, both ousted for going too far, complaining about Anthropic.

> “We have a similar situation in mobile where Apple may not be considered a monopoly, but people have walked around for a decade with a supercomputer in their pocket that is wildly underused.”

In that we can’t run a Torrent client to download illegally redistributed media 99% of the time? Otherwise, in what way, are they underused? For the degrees of public addiction, a more underutilized phone would be a social benefit.

Let me back up what you're saying. They absolutely are not a monopoly today by any definition, by any stretch, in any conceivable way.

I'm looking forward. Things are moving very quickly. As I said above, I'm afraid of us diverging into another Apple situation in the future. If I suggest that they should be looked at and thought about, it's not for today, it's for tomorrow. If divergence continues. Because as with everything in AI, it might hit us a lot faster than people expect. Hell, given their approach to morality, I suspect that Anthropic folks have already thought deeply about these sorts of concerns. That's why it's actually a lot more in character for them to be doing this not due to self-preferencing, but due to unaffordability, which - if you look at my first post - is what I said seems to be happening.

Suffice to say that I have a graveyard of things that I think phones could have been, where unfortunately we've ended up with these - as you say - addicting consumerist messes.

Gonna stop here so I don't flood the thread. We're getting very off topic.

OpenAi is burning cash to stay relevant aiui, i.e. they will keep subsidizing

You can use these tools with most providers today, just no subscription plan. If you have enough spend, you can likely get bulk deals

Doesn't OpenAI allow this today?

It's a good way to win market share and build goodwill, but one has to wonder whether this class of usage is marginally profitable for them (or anyone) and how sustainable their lenient policies will be for them long term.

You mean whether another competitor will emerge? Right now we have OpenAI.

The real threat that Anthropic sees as real competitors in the long term, are the AI labs building open weight models, especially the AI labs in China.

Some of the Chinese labs with cheaper per token costs do support it, like say minimax: https://agent.minimax.io/max-claw

I haven't tried it to see if it's any good but it's $20/mo.

Kimi seems to support this with their 39 usd a month plan.

I agree, eventually the open models will be good enough and we can pay for our own infra and cut out the middle man. Also, the smaller frontier are nearly as good today and I expect the mega models will be used primarily for distillation

What you're saying is conceptually true for subscription services in general, but thats not why they are making this change. There's a 5 hour limit and a weekly limit. Those are hard token limits. Everyone on a plan pays for the max set of tokens in that plan. The limits manage capacity. The solution to that isn't a change of ToS, it's adjusting the limits.

In other words this is about Anthropic subsidizing their own tools to keep people on their platform. OpenClaw is just a good cover story for that. You can maximize plans just as easily w/ /loop. I do it all the time on max 20x. The agent consuming those tokens is irrelevant.

For what it's worth I don't use OpenClaw and don't intend to, but I do use claude -p all the time.

You aren't paying to be using that limit all of the time.

You are paying to be using that limit some of the time. There are 5 hour windows when you are sleeping and can't use it. There are weekend limits.

Theoretically you can max out every 5 hour window, but they lose money on that.

It's structured so users can have bursts of unlimited usage, and spend ~15% of the theoretical max cap, and that's still cheaper than a subscription for that user.

An OpenClaw user can use 6, 7, 8 times what a human subscriber is using.

How can an OpenClaw user use 6 times what a human subscriber is using when I'm four hours into the week and 15% of my weekly limit is already used up, just by coding? OpenClaw can't use 600% of my weekly limits.

>How can an OpenClaw user use 6 times what a human subscriber is using when I'm four hours into the week and 15% of my weekly limit is already used up, just by coding?

Perhaps because your Claude agent usage is not representative of the average user, and closer to the average OpenClaw user levels...

Without data, this is just a bunk excuse to defend the walled garden practices.

With data, it's an engineering target.

They could just 429 badly behaved clients.

They already 429 everyone! That's the crazy thing. They already have strict limits that we all keep hitting regularly.

Man, I run 3-5 sessions an evening for 5-6 hours, and longer on weekends and feel like I'm barely using what I paid for. I've only hit five hour limits a small number of times. Genuinely baffled when I hear people blow through tokens apparently several times faster than me. Are you going out of your way to design complex subagent workflows or something? I just let claude code use subagents when it wants to but don't give it any extra direction to use them.

Not sure what tier you're on.

Basically; spin up in the morning eats a lot of tokens because the cache is cold. This has actually gotten worse now that Opus supports a 1Mt context.

So: compact before closing up for the night (reduces the size of the cache that needs to be spun up); and the default cache life is 5 minutes, so keep a heartbeat running when you step away from the keyboard to keep the cache warm.

Also, things like web-research eat context like crazy. Keep those separate, and ask for an md report with the key findings to feed into your main.

This is not exhaustive list and it's potentially subtly wrong sometimes. But it's a good band-aid.

https://news.ycombinator.com/item?id=47616297

Know what's funny? Openclaw might actually burn less tokens than a naive claude code user; if configured correctly. %-/

I'm on the $100 tier, but I don't use OpenClaw. My point is it can't use more than 100% of my limit, so "6-8x more" is only possible if you use 15% of your subscription normally.

Right. I was editing to add more info. Possibly you already know the usage tricks I list above. The world is still very messy and not much is documented properly.

And I'm skeptical of the 6x-8x claim myself. They'd have to explain that in more detail.

Oh actually the cache trick is neat, I hadn't considered it, thanks!

Efficient token use will be the new code/vim golf.

Whether it's human token use, or future OpenClaws

I've mention before that we should have a look at Telegraph/telegram speak. There was a HUGE industry in word efficiency at that time. There are hundreds of books.

I even think an LLM trained to communicate using telegram style might even be faster and way cheaper.

> I've mention before that we should have a look at Telegraph/telegram speak.

.- -. -.. / .. --..-- / ..-. --- .-. / --- -. . --..-- / .-- . .-.. -.-. --- -- . / --- ..- .-. / -. . .-- / - . .-.. . --. .-. .- -- -....- -... .- ... . -.. / --- ...- . .-. .-.. --- .-. -.. ...

.--. .-. .- .. ... . -.. / -... .

Why use many word when few do trick?

Reminds me of the terminus agent/harness on the terminal-bench coding benchmark - they just send send keystrokes to a tmux session. They score pretty well.

https://www.tbench.ai/news/terminus

It’s the new cloud cost vector, where cutting 2K from context on a busy service saves $xxxxx.

Terse.

No org doing real work cares about token use costs.

This mainly just affects hobbyists.

Token use cost can easily get as large as dev salaries. Even real businesses care about that.

Like "Token Usage Consulting" companies popping up now? :-D

> You aren't paying to be using that limit all of the time.

The erosion of the norm of things doing what they advertise rather than being weasel-worded BS is particularly unfortunate, and leads to claims like this.

I feel like Anthropic is going down a bad path here with billing things this way. Especially as local LLM continues to develop so fast.

I downgraded from my $200 a month plan to my $20 plan and hit limits constantly. I try to use the API access I purchased separately, and it doesn't work with Claude Code (something about the 1 million context requiring extra usage) so I have to use it Continue. Then I get instantly rate limited when it's trying to read 1-2 files.

It just sucks. This whole landscape is still emerging, but if this is what it's like now, pre enshittification, when these companies have shitloads of money - it's going to be so much worse when they start to tighten the screws.

Right now my own incentive is to stop being dependent on Claude for as much as I can as quickly as I can.

This is how free drink refills, airplane tickets, Internet service, unlimited data plans, insurance, flat rate shipping, monthly transit passes, Netflix, Apple Music, gym memberships, museum memberships, car wash plans, amusement park passes, all you can eat buffets, news subscriptions, and many more work.

Either you get a flat rate fee based on certain allowed usage patterns or everyone has to be billed à la carte.

This is a different case - those all have limitations based on human behavior (it's not necessary or possible to constantly be washing your car the entire month when you pay for unlimited washes) - that doesn't exist here. The types of plans available should reflect that reality. If gyms faced a situation where people would go and spend 18 hours working out every day for a month, they would probably change how they billed things.

Your comparisons are all also "unlimited" situations to Claude's very much limited situation. You can't buy a plan for Claude that is marketed as being unlimited. They're already selling people metered usage. They're just also adding restrictions on top of that.

They sell metered usage while having the implied expectation that most wont use it fully. Power users and users of stuff like OpenClaw don't match that idea.

So they further restricted the metered caps, which were only offered to NOT be reached by that many.

Simple as that.

>Power users and users of stuff like OpenClaw don't match that idea.

Then they should figure out how to structure an offering that accommodates this type of usage not just blanket ban it

Why "should" they? There's no reason they would especially when their competitor now owns OpenClaw.

Because a big part of Anthropic's story is that they build based on how people actually use AI. Power users aren't just annoying edge cases, they're signal. Throttling them and calling it done is inconsistent with that.

Sorry to tell you but generally any company's "story" is all marketing and PR, if it interferes with their making money, which it does in this case, that company will not hesitate to leave it behind.

Oh the billion bollar vc backed pre ipo companys story was this? Omg and they somehow are not delivering up to your standards? Damn they better get their act together lest people like you will whine on twitter about them losing their way

> Power users aren't just annoying edge cases, they're signal.

You got that right; in this case they are signalling that AI token providers are not going to be able to run at a profit anytime soon.

Not sure if that helps or hurts your argument, though.

> Power users aren't just annoying edge cases, they're signal.

Not all power users. Some re-invent the wheel and/or do things inefficiently, and in most cases there's no business incentive to adapt the service to fit the usage patterns of those users, or of other users that deviate from the norm in regards to resource usage.

They build based on how _people_ use AI.

> Why "should" they?

Because it is clear that there is a market demand for it.

There is also a clear market demand for $10 bills sold for $5, but I don't see you tapping into that opportunity!

I didn't write anything about pricing. I just claim that people would love an offering without the discussed restriction, and because there is clear evidence of such a demand, it would make sense for Anthropic to prepare such an offering.

>I didn't write anything about pricing

Yes, and that's exactly the problem I'm pointig at.

Your comment "that people would love an offering without the discussed restriction" ignores the pricing burden of that, which is why it's confused why Anthropic don't just offer this.

> I didn't write anything about pricing. I just claim that people would love an offering without the discussed restriction,

The API has no restrictions; what is the people's objection to that?

Then you mean the API, and if that's not sufficient, then you do have an issue with wanting something for nothing.

They did figure out how to structure an offering that accommodates that type of usage: pay for your tokens.

They did: just use the metered API.

> Then they should figure out how to structure an offering that accommodates this type of usage

They did, didn't they? You can pay the non-plan rate.

> not just blanket ban it

They didn't do that. The email specifically tells you how to use Openclaw with Anthropic. There is no "blanket ban".

Don’t cry while you’re ruining it for everyone.

Isn’t that just usage based charges?

>If gyms faced a situation where people would go and spend 18 hours working out every day for a month, they would probably change how they billed things.

Not the best example. The upkeep cost of a gym is pretty flat regardless of how much people use the facilities. Two people can't use a single machine at the same time make it wear out twice as fast. The price of memberships is not correlated to usage, it's inversely correlated to the number of memberships sold.

>Two people can't use a single machine at the same time make it wear out twice as fast

The machine doesn't care about the number of people using it. If it's constantly being used, it will wear out faster. You are conflating "we price based on expected under-utilization" with "costs don't scale with usage." Those are different things.

The inverse correlation you talk about isn't relevant here - People buy gym memberships intending to go, feel good about the intention, and then don't follow through. The business model is built on that gap. That's pretty specific to fitness and a handful of similar industries where aspiration drives purchase.

Anthropic doesn't sell based on a "golly gee I hope people dont use this" gap - they sell compute. Different business.

>If it's constantly being used, it will wear out faster.

Yeah, but there's an absolute limit to that, beyond which the cost doesn't keep increasing. Beyond that point, the QoS goes down (queues).

>You are conflating "we price based on expected under-utilization" with "costs don't scale with usage."

I'm not conflating anything, I'm responding to what you said:

>If gyms faced a situation where people would go and spend 18 hours working out every day for a month, they would probably change how they billed things.

Why would a gym need to change how they bill things if all their customers were aiming for maximal utilization, when their costs would barely see any change? I doubt your typical gym operates on razor-thin margins.

Gym costs absolutely scale with usage. Equipment wears faster under heavier use. Cleaning and maintenance staff hours scale with how much the facility is used. Consumables like towels, soap, and chalk go faster. HVAC runs harder. The reason gyms can offer flat-rate pricing is that they bet on under-utilization, not that costs are flat.

Setting that aside, even if we accept your argument that gym costs barely scale with usage, then that makes gyms a bad comparison case for Anthropic, whose costs directly scale with usage. You can't use the gym model to defend Anthropic's pricing decisions if the two cost structures are nothing alike.

I'm arguing that both gyms and Anthropic have usage costs that scale with usage, but gym business model assumes a large margin of under-utilization and there's a hard cap to "power user" - I think both of those extremes don't apply to Anthropic's situation. Under-utilizers aren't paying for AI they have a free tier. There's also a natural ceiling on how much any one person can use a gym. There's no equivalent constraint on API usage.

>You can't use the gym model to defend Anthropic's pricing decisions if the two cost structures are nothing alike.

Am I? I think you read something into my comments that I didn't write.

> The reason gyms can offer flat-rate pricing is that they bet on under-utilization, not that costs are flat.

Yes. In fact i remember hearing about a gym which offered a flat-rate pricing model but explicitly excluded certain professions from partaking in it. I remember the deal was excluding police, bouncers, models, actors and air stewardesses. They had a separate more costly tier for these people. (And I think i heard about it from the indignation the deal has caused online.)

> Under-utilizers aren't paying for AI they have a free tier.

Sure they do. Free tiers suck. I may not always need to use AI, but when I need it, I don't want to immediately get hit by stupidly low quotas and rate limits, or get anything but SOTA models.

> Anthropic doesn't sell based on a "golly gee I hope people dont use this" gap - they sell compute. Different business.

There is nothing anywhere hinting at that.

They don’t sell compute. They sell a subscription for LLM token budgets that they hope people don’t use because the compute is vastly more expensive than what they charge or what users are ever willing to pay.

Especially with enterprise subscription plans the idea is for customers to never utilize anywhere close to their limits.

Two people can't use a machine at the same time is the issue. If you have 50 machines and 200 customers all of whom want to be in the gym 18 hours per day that's quickly going to lead to cancelled subscriptions. Now you need more space and machines or some other way to balance things.

Agreed, but it's an indirect causal link, not a direct one. If the demand far outstrips the possibly supply the demand will have to go down, and it can either go down by people accepting that they can't be in the gym as much time as they would like, or as you say by memberships being cancelled (in which case the price may go up or something else might change).

"Unlimited" has always been a lie. There is no free lunch. There are always limits.

I've had to unwind "unlimited" within startups that oversold. I've been bit by ISPs, storage providers, music streamers, fuckin _Ubers_, now AI subscription services, that all dealt in "unlimited". None of them delivered in the long run.

I'd be mad at Anthropic if it weren't for the fact that my experience now can see this sort of thing from a mile away. There are a lot folks, even on HN, that haven't been around for as long. I understand the outrage. I've been there. But these computers cost money to run, and companies don't operate at a loss in the fullness of time.

Once you know that unlimited trends towards limited, the real question is whether we're equipped as a society to deal with the fact that the capital-L Labor input to the economic equation is about to be replaced with a Capital input for which only a handful of companies have a non-zero value.

On your 1.5Mbps link, you could theoretically download 500GB per month. A huge amount, but I believe it was often genuinely allowed, because their uplinks could cope with it. Unlimited could genuinely be unlimited.

But now you might get things like “unlimited” 1Gbps… which reverts to 10Mbps (1% speed) or worse after 3.6TB (eight hours). And so your new theoretical maximum is about 6.8TB per month rather than 330TB.

You can both know that "unlimited" means "limited" and also be pissed that they market it as such and try to conceal the actual limits.

> You can both know that "unlimited" means "limited" and also be pissed that they market it as such and try to conceal the actual limits.

Reminds me of when ATT had a fake 5G decoration on phones.

"AT&T won’t remove fake 5G logo even after ad board says it’s misleading"[0]

You can just get away with lying. That's the level of enforcement that exists against unethical behavior in business today.

0. https://www.theverge.com/2020/5/20/21265048/att-5g-e-mislead...

This is all just the classic "the first hit's free" business model.

The commons? Tragic.

à la carte is honest; overprovisioning just slows progress by preventing demand from creating pressure to innovate proper solutions.

Rent doesn't work that way... yet. Imagine if it did though, people would be arguing:

"Well, you're not expected to be able to live in that home the entire month that you paid for!"

Internet service and unlimited data plans work like this though.

>Especially as local LLM continues to develop so fast.

I'm sorry is there anything even close to sonnet, much less opus, that can be run on a 4080? Or 64gb of ram, even slowly?

Qwen 3.5, Gemma 4

Opus 4.6

Look for the current crop of local Mixture of Experts models, where it seems like they've made inroads on the O(n^2) context attention cost problem. Several folks have mentioned Qwen, but there's many more of that ilk. Several of them actually score really high on benchmarks. But when I mess with one of them locally by hand myself, (I have a 3090), it feels a bit like last year's Sonnet. They don't quite make the leaps of understanding you get from Opus.

* Weird thing of the day: https://huggingface.co/Jackrong/Qwen3.5-27B-Claude-4.6-Opus-...

You can run SOTA local MoE models very slowly by streaming the weights in from a fast PCIe 5 SSD. Kimi 2.5 (generally considered in the ballpark of current sonnet, not opus of course) has been measured as 2 tok/s on Apple M5 hardware, which is the best-case performance unless you have niche HEDT hardware with lots of PCIe lanes to attach storage to and figure out how to use that amount of parallel transfer throughput.

Well, I reinstalled LM Studio today after some ~10 months since I last used it, just to test Gemma 4. On my PC with 32GB RAM and 4070 Ti (12GB VRAM), it (Gemma 4 26B A4B Q4_K_M) loads and runs reasonably fast, with no manual parameter or configuration tuning - just out of the box, on fresh install - and delivers results usable results on the level I remember expecting from SOTA cloud models 12-16 months ago. And handles image input, too. I'm quite impressed with it, TBH. It's something I can finally see myself using, and yay, it even leaves some RAM and VRAM left for doing other stuff.

And the smaller Gemma 4 models can do audio too.

The Qwen models are also really good.

A ~$5000 USD Macbook can run open source models that are competitive with GPT 3.5 or Sonnet 3. So on nice consumer hardware you can have the original groundbreaking ChatGPT experience that runs locally.

You can use the API with CC, you just need to log out and log in, selecting API usage.

> I feel like Anthropic is going down a bad path here with billing things this way.

What do you expect them to do? You are looking at a business currently running at a loss, and complaining about their billing even though this is not a price-rise?

Unrelated, is it still possible to use $10k/m worth of tokens on their $200/plan?

They seem to know what they’re doing. Anthropic entered 2025 with a run rate of $1 billion; the run rate for March 2026 is estimated at $19 billion.

Internal projections show the company reaching cash-flow break-even in 2028, after stopping cash burn in 2027.

They’ve already implemented several of the features that put OpenClaw on the map.

> Anthropic entered 2025 with a run rate of $1 billion; the run rate for March 2026 is estimated at $19 billion.

I don't know what that means in this context.

> Internal projections show the company reaching cash-flow break-even in 2028, after stopping cash burn in 2027.

What does that have to do with them implementing restrictions on their plans because they are currently running at a loss?

Okay, lets say their internal projections[1] are accurate: were those before or after Openclaw released? Maybe their projections were made on the assumption that people would stop using $10k/m worth of tokens on a $200/m plan? Or that those users doing that will only be doing code? Or that the plan users won't be running requests at a rate of 5/minute, every minute of every hour of every day?

--------------------------------

[1] Where did you find those projections? I'm skeptical, at their current prices and current plans, that a break-even at any point in the future is possible unless they shut off or severely scale down training. Running at a per-unit loss means that the more you sell, the larger your loss - increasing your sales increases your loss.

If you can do less for the same price, that is in effect a price increase.

Use qwen 4.6 plas

We can hope that they optimize the models. I still think its going to be very hard for them to charge $100 or $200 a month at scale from many people, especially with AI "taking jobs". To the extent that happens most of those people won't find replacement income.

> Theoretically you can max out every 5 hour window, but they lose money on that.

Then it's not priced correctly. As I said, you can do all of this without OpenClaw.. claude code ships with everything you need to maximize the limits.

It is priced incorrectly, but that is intentional. You can't create a tiered paid plan for the whole world that fits everyone. You can't create nuanced extra plans to satisfy all the outliers. It's an bet to keep the customers and still having a good margin. Think of ecom, returns are a big struggle for any large company because they are unpredictable and subject to abuse, shipping fees are just an sophisticated guess to cover that cost. Not a subscription, same mechanics. The only thing here to criticize is, if it's a good thing to make everything a subscription and disguise the real cost.

>You can't create a tiered paid plan for the whole world that fits everyone.

I mean, you can. Electricity is already sold that way. Subscribers with uncharacteristic usage spikes don't get blackouts, they get a slightly larger bill, and perhaps get moved up a tier.

Very valid. My comment was fixated around the fact that big tech has the addiction to have subscriptions for everything. It's common that you provide generic subscription plans for the masses and supply "call us" custom plans for the specific (usually corporate) needs. If anthropic doesn't provide that or vibe coders are too cheap to do that, then those are issues, but the subscription models are itself valid. It is certainly misleading to a degree, but we've stopped complaining about this a while ago.

It's pretty stupid because as others in this thread have pointed out it's already not a flat plan. Even from their side it makes zero sense to bill things this way rather than based on usage. It's not like a VPS where your VM shares the hardware, which consumes electricity more or less regardless of what you use the machine for.

Those yottabytes of VRAM are also consuming electricity constantly.

The difference being that an LLM request is not an operating system. Since they're compartmentalized and ephemeral, you can very easily distribute requests among your available hardware so that you can switch off machines during periods of low activity.

Your capital costs for buying those machines don't go away.

That's a problem that already exists in power generation and delivery, and it's already been solved. Bills are sums of fixed terms and variable terms.

Custom payment schemes are late stage profit generation. It requires hoards of salespeople or an AI that can actually do math.

It's just how hyperscaling works. You are not wrong, but in the wrong timeline.

I'm not talking about custom, negotiated service contracts, I'm talking about simply charging people for what they use.

But that would be using (a special Claude code version) of the API; as it stands now, I have tried the current api for fun and I hit $200 well within an hour. So if they would charge for real use, no one would use it as there are competitors who have less harsh limits with tier plans still. If all go away then I will be running open models on vast.ai or so as those are now viable (been testing with glm 5 and it's great for coding). So tier subscriptions cannot go away as it will end those companies fast.

No, it is priced correctly.

Just because outliers can be money-losing doesn’t mean you should raise the price for everyone.

> Just because outliers can be money-losing doesn’t mean you should raise the price for everyone.

If they are losing money then it's not priced correctly. That's what I responded to.

Yes, subscriptions work as you say. Plenty of people under utilize subscriptions from prime, to credit cards, to netflix. But if they lost money overall, they too would raise prices. Because that's how economics works. Shortage of capacity, high demand, raise prices until equilibrium.

There's other knobs beyond ToS. They just didn't choose those options.

Yes, they chose the knob of ToS, because that was the way to price it correctly.

The market will determine if it was the correct choice. I don't think it's an obviously bad choice on their part.

> If they are losing money then it's not priced correctly.

Just a few years ago this was the standard business model for startups: attract VC money, offer plans at a loss, capture a huge market, boil the frog with incremental price increases to become profitable.

Companies like Uber wouldn't have been anywhere near as successful if they had been forced to make a profit from day one.

I think "offer unlimited but ban people who are extreme enough outliers" is actually a pretty normal way that things work instead of "raise prices until equilibrium is reached", including in credit cards. Credit cards do simply ban people they think are "rewards churning" based on a completely subjective TOS policy for example.

Raising prices is a bad strategy if you have a smaller base that costs enormously larger. "A million users that cost $1 and one user that costs $10 million, charge everyone $10 equilibrium", you're screwing over almost all of your users. The $20/month sub price is basically just not trying to capture the openclaw users.

Yes, it's an unsurprising strategic choice. It's just sloppy PR that places the blame on OpenClaw somehow being irresponsible, when the actual rationale has little to do with that.

Usage will be roughly a bell curve. Losing money on the extreme upper end is NOT a reason to raise prices for everyone, because that will decrease overall revenue (see: price elasticity).

ToS is the right knob here. When 0.05% of your customers use your product in a way that increases your costs 10,000%, you find ways to stop that. You don’t amortize the cost increase across everyone else.

Yes, and they are in control of Claude Code, so they are fine with that. If it causes problems they can tweak it. If OpenClaw causes problems they can’t.

you can write automated MCP tools that run within claude code, and could theoretically generate as high a load as any other automated/3rd party agent. You can also do loops that burn tokens incredibly fast. This is allowed with no caveats (I use MCP's basically to test what I'd like to try with the API...) So this explanation just seems a lil hollow.

When you can’t enforce everything at once, you go where the most acute problems are. I imagine when your MCP avenue of abuse catches on—like this other category of harnesses did—to such a scale as to become a problem impacting us folk trying to go about our business… when that’s where the problems shift, I imagine (and hope) Anthropic will crack down on that vector too. To keep the service usable for us ordinary meatbags.

I’m glad they give us the leeway to experiment, and I’m also glad they weed the garden from time to time. To switch metaphors, I’m deeply frustrated when my very modest, commuter-grade use gets run off the figurative highway by figurative hot-rodders. It’s been extra-529y this week, and it’s about time they reined it in a little.

You’re always welcome to pay-as-you-go for as many tokens as you’d like to burn on their infrastructure… or to compute against any of the wide array of ever-improving open models on commodity compute providers…

>>when your MCP avenue of abuse catches on

Thats an interesting way of phrasing it - so is there a way to use the quota that's not 'abuse'? MCP/claude code seems to be want they want you to use it - are loops or ralph abuse as well ?

It's not difficult at all to burn through your weekly limit just writing code.

I take your point, the way I used “abuse” there probably carries more charge than I’d meant it to. And I agree: Anthropic clearly want people to industrialize and automate usage. But that’s not what the subscription product is for. Use all the loops you want, burn all the tokens you want—just pay what they cost.

> is there a way to use the quota that's not 'abuse'?

I think my answer is “no.” In that I’ve never thought of the limits as “quotas,” and I don’t think I’ve heard Anthropic speak of them that way. Quotas are to be used up, while limits are to signal that what you’re doing is outside the envelope of acceptable use. Quotas are to be met, limits are to be avoided.

I interpret the intention of the subscription, like a membership at a makerspace, to be to allow novices to experiment with stuff, to allow them to learn without having to understand the economics upfront. To play without fear of expensive mistakes. And, like the makerspace, it can only offer generous limits because most of us will rarely bump up against them.

If these harnesses or OpenClaws or whatever stop making sense as soon as they have to pay their actual costs, then that’s a pretty good sign they’re abusing the spirit of the subscription.

But Anthropic seem more than happy to service those uses via the API or metered usage, and even to sweeten the deal with more reliable access and bulk discounts.

>>I take your point, the way I used “abuse” there probably carries more charge than I’d meant it to.

Fair enough.

>> But that’s not what the subscription product is for.

This was the point I was trying to make - I pay for XX tokens/usage. But somehow using them all is 'taking advantage' ?

BTW - I'm actually not complaining about the limits - I probably only use half my tokens on average week. I'm just annoyed at having to jump thru hoops if I want to try something 'API' oriented. For me, AI is still the new shiny - I try all different sorts of things learning/playing. There was an article posted today about writing agent harnesses. That could be interesting - maybe I want to try my hand at it. But then I've got to mess around/pay extra to _try_ something I that my subscription already easily covers.

[added:] >>to take on personal-scale projects, to allow them to learn without having to understand the tool’s economics upfront. To play without fear of expensive mistakes.

This is exactly what I'm trying to do - however, as soon as you want to try anything 'API' oriented, the 'fear of expensive mistakes' comes right back.

Yes, but very few people are actually doing that compared to OpenClaw. If everyone else was doing that, they'd be cracking down on it too.

Anthropic is much more concerned about what people are ACTUALLY doing than what they could, in theory, be doing.

While you can write an automated tool to consume all their tokens, I strongly suspect most users, like myself, are not doing that. So even if Anthropic loses money on a power user, they profit overall and keep public sentiment high by not alienating users with restrictions. It's an optimization problem of making a profit off the average used while staying low enough to attract customers, even if that means some users cost more than they pay.

More users spinning up OpenClaw means that balance starts to shift towards more users maxing their tokens, thus the average increases, so I think their explanation makes sense still.

>>So even if Anthropic loses money on a power user, they profit overall and keep public sentiment high by not alienating users with restrictions

So they profit overall if I use all my tokens either way? Again, I understand usage limits - I just don't understand why some usage is 'good' and some 'bad' if I'm using the same either way.

>>More users spinning up OpenClaw

I'm pretty sure that's a small percentage of overall users, and probably skewed towards the very people that would be recommending/implementing you model for work/businesses. Seems like that would be the group you are encouraging/cultivating ?

My company has several MCPs that our very token intensive, but it seems that with Claude Code, usage is throttled even before hitting limits. I don't have any proof, but often when using intensive MCPs, Claude Code will just stall for 10+ minutes.

I wonder if anyone else has experienced this?

I think maybe you are not familiar with what /loop and the Claude cron tools do.

https://code.claude.com/docs/en/scheduled-tasks

I need a hypothetical use case for things like this, I don't get how so many people have so much desire for use of features like this.

https://martinfowler.com/articles/harness-engineering.html it's being talked about everywhere.

If you manage developers or product folk, do you allow them to work when you're not looking over their shoulder? All developers can be managers/team leads now. You plan, you delegate, you review.

You're welcome to not do this, surely that's appropriate in quite a few areas of work, but many of us are because we can get more work done than if we we're micromanaging every line of code change. For startups, where a bit of quality can suffer in favor of finding market fit, this is huge.

The actual /loop and cron are beyond the normal "agentic loop"

I'm not sure what point you're trying to make. `claude remote` on a secure vm is basically all you need to operate a factory from your phone. I suspect a lot of people with your line of reasoning are stuck on human in the loop while awake level of AI use. Anthropic has no interest in that long term and all of their product moves validate that.

Every morning it summarizes a bunch of stuff for me, suggests me PRs to review, emails to reply to, freshly cloned any new repos, pulled all others, presents me with the suggested approaches to my PRs of that day, and gives me a list of my slack mentions that look more urgent.

This is just the morning ones, and saves shitloads of time of clicking around from tool to tool, freeing up time for the thinking and deciding.

Wow, you should probably ask it to write a script for 90+% of that instead . Sounds like a huge waste of electricity.

How? Most of what was mentioned requires discretion and judgment. You can question whether an LLM would be able to offer that, but there’s no script that can do b it.

Deduplicating/validating/processing incoming bug reports?

I think maybe you still don’t understand that not everyone will max out their usage, regardless of the methods available.

People don't pay double the $100 account in fixed recurrent payments if they don't intend to use a lot more than they would use in the $100 subscription.

Perhaps people at Anthropic should ask Sonnet (or Kimi, it's much better value) how power laws and pareto distributions work? You are advertising for people who can justify a virtually unlimited amount of tokens, why is it surprising that they would use as many as you're offering them in the plan?

PS: interesting that you'd use a throwaway account to post this

> Theoretically you can max out every 5 hour window, but they lose money on that.

No, there is a weekly limit as well. Maxing out a single 5h window uses ~10% of the weekly limit

I fill my week limit in a few days :(

Do you do process XLS or similar data? Ive seen using any other files than code files eats much more faster

now maxing out a 5h window takes it to 20%. just experienced it today. so clearly they have reduced the subsidy on plans along with this

I've met people that fill a box of sushi to take home at the end of their “all you can eat” session because “they paid for it”. Shrug.

Yes and the staff will tell them to stop that or charge them extra for it.

Yes and they will hide their sushi-grabbing because somewhere deep inside they know it's not part of the deal, while at the same time still strongly feeling that they have indeed paid for it.

Ah, to be human!

I'd argue they hide their takeaway because of what GP comment said — not because of anything innate, but because a staff member will not let them.

I grew up in an Asian household of six. We definitely took food home at AYCE places. My parents definitely knew it wasn't OK, but they felt like they were gaming the system (like a dubious life hack of sorts) and saving money, so they were actually quite proud of it, bragging to friends how much they were able to get.

To be human indeed!

I assume it's not unusual for thieves to brag about their scores.

Ah to be inhuman!

People brag about all sorts of things moral and immoral.

Calling them thieves is a bit harsh, it's not like they didn't pay for the food, just not able to transport it unless it's in your own internal containers.

Yes sorry, in case it wasn't clear, I wasn't agreeing with the commenter or calling my family thieves :) just because a restaurant kicks you out because you took too much food doesn't mean you're a criminal.

In the Eastern Bloc states, it used to be so common for workers to steal from the workplace that new moral norms were established around this; if you're not stealing from work, you're stealing from your own family!

Goes to show just how fragile a high-trust society is. Theft and corruption can easily be normalized to such an extent that not participanting gets reframed as immoral.

Wow, I was just saying to a friend that I couldn’t understand people risking their jobs to steal stationery or toilet rolls from the workplace.

I guess if it’s your moral obligation to steal from the workplace it reframes it somewhat.

The slogan of the Russian Revolution of 1917 was: "Factories to the workers, land to the peasants."

If the factory is yours, then everything inside is yours ;)

But it's funny how low wages under the broken Soviet economic system turned such things into a semi-official, informal work perks, allowing people to make ends meet.

This is not false but totally an oversimplification.

I don't think communism is a good form of government and I don't think the soviet union was marching the right way.

But the biggest blunts came from other much more serious mistakes caused by politicians ignoring science, like the big famine and many others, including the Chernobyl connerie

It was less the low wages and more the general unavailability of things (shortages). Lots of things you couldn't just buy but you had to know somebody who knew somebody.

I wouldn't call it "funny" though. It ws quite sad and I'm glad it's over.

As I mentioned in another commebt, I don't even consider anything related to that to be a viable government system.

That said, the general unavailability of everything was caused by an incompetent government rather the the system itself but the system itself caused the government. My point is that it was a succession of demagogueries hiding personal interests that caused the recurring and unrecoverable tragedies of that state. Being controlled and misguided is not exclusive to any particular government or political system.

System consistently produced an incompetent government that had lots of power.

The general unavailability of everything was caused by the system.

Hmmmmm

In Poland, factories' light installations ran on non-standard voltage, so that, if you stole the lightbulb, it'd be useless in your household. Unfortunately, they couldn't figure out similar solution for toilet paper and many other household items.

As for motivation, beyond the obvious one, people also stole because many items were not available in stores (having a guy who has a connect on toilet paper was a thing back then), and also, since Communism in Poland was actually a Russian dictatorship, the idea was that if you steal from communist factories you're fighting the system and making it fall faster.

It is not a justification, but, it is not like Anthropic didn't pirate tons of books and burnt evidence... The only difference is that books don't have a terms of service

Just curious, how does it work? Did you head to the restaurant with tupperware containers and hid them in bags or purses?

Zip lock bags and oversized purses

Ha, that's awesome! I am not judging, I totally can see some of my family members doing that, too.

The cooperative and competitive sides of our soul fighting it out in a single situation

Really

Tell me you are not using Anthropic without telling me. Bursts of unlimited usage was never the case. And I bet their infrastructure doesn’t like bursts as much as more spread out activity.

> Theoretically you can max out every 5 hour window, but they lose money on that.

This typically results in a ban for TOS violations after a few windows in a row on a claude subscription

I have maxed out my 5 hour limits and my weekly limits fairly regularly, when I did a bunch of editing work on long form writing next to having CC run a few coding tasks over the xmas holidys - I only slept a few hours at night an timed those roughly (by chance) with my limits.

I neither got a warning or a ban or anything - and that was with the double token amount during those days.

So I don't see human usage being something they ban for TOS violation, like you describe. But as always YMMV.

Was this on a new (few months) or significantly older account?

I am on a pro Plan for I don't even really remember and on a Max plan for I feel about 7 or so months.

It wouldn't really surprise me if their automatic systems are, one way or another, looking for that sleep period. Those ostensibly human users who use the service with no breaks for sleep would be naturally suspect.

What are they going to ban you for? Using their products? If Anthropic didn't want you to use their LLMs around the clock they wouldn't publish the features on their consumer app to do just that.

[delayed]

> You are paying to be using that limit some of the time. There are 5 hour windows when you are sleeping and can't use it. There are weekend limits.

They could easily structure their limits to enforce that kind of pattern fairly on both human and automated users. They could e.g. force a cooldown period between your daily activity bursts, by decreeing that continued heavy use on a 24h basis would count exponentially more towards your limit. That would be transparent and force the claws to lighten their load below that of a typical human user. We're talking about a company that's worth hundreds of billions of dollars and targeting highly sophisticated enterprise users, not consumers; it's just not credible that they'd be technically unable to set that up.

Or alternatively just pay based on what you use? I.e. $/tokens.

They offer that, API pricing. Or "extra usage" which they're saying you can still use for OpenClaw. It's really expensive!

Everyone want to take their subsidies from the consumer subs and use them to avoid paying the real costs.

Even the API pricing is subsidized by investors. When that stop, pricing will escalate.

Is API pricing subsidized on a unit basis?

Most estimates I've seen have shown that API usage seems to be at least unit profitable (paying for infra and electricity, not R&D)

I believe it was 2 years ago. But costs have risen a lot in the last couple of years but prices have decreased with competition.

I don't believe that Anthropic looses money when heavy users consume the max amount of tokens.

do you have any proof of your statement ?

Obviously, Anthropic is a private company, so nobody here is going to know their financials (who is at liberty to share them). I'm not GP, but I think it's reasonable to assume the subscription is priced based on average usage, and that's a major reason it's so much cheaper than API prices.

If we do same work via Subscription and API , the subscription is way cheaper. So if we compare them yes they loose money.

none of them are making any money yet. they all lose.

At least on a personal max account, I can't max every window. There is also weekly limit. If I max every window, I run out of tokens halfway through the week.

I think the gp understands that, he is stating that openclaw (has cron that runs every 30 seconds) will use up the last drop of juice the plan offers - aka ultimate power user.

And the point being made is that can be done using only Anthropic provided tools. Ask claude to set up crontab for you and be proactive. It will happily do so. This spring/summer you'll find all the same stuff they OpenClaw has in a Claude product. Most of it is already there just not packed neatly for a non technical user.

And it sounds like that will require additional fees.

> You are paying to be using that limit some of the time.

This makes zero sense. I'm paying to use that limit all of the time. If that's too much for Anthropic, they are free to lower the limits or increase the price. Claiming otherwise would be false advertising.

They did? What do you think that email to the user was about?

They lowered limits opaquely before this. They "announced it" in a twitter by a tech lead. This time it was in an email on a Friday to only some customers.

If you maximise the usage of your quota you are not doing anything wrong. They just tricked people I to thinking they quota was higher than it was really was and when people found a way of maximizing that, they had to cut it.

You guys are arguing on the reality of a subscription, but Anthropic still resides in the coocoo make-up world of growth at all costs backed up by unfathomable investments. They're not acting rationally by trying to present a good product with reasonable backend fundamentals. They're just trying to maintain the money loss to what they have set aside for the quarter. OpenClaw was not planned for, and thus must be fought.

Anthropic isn't "fighting" OpenClaw. They just want OpenClaw users to switch to API pricing so that their service doesn't become a blackhole for investor money. Operating at a loss can be strategic, but they had to carefully consider the ratio of casual users to power users to keep that loss steady and sustainable.

Power users always cost these services more than they pay, and OpenClaw turns every user into a power user. A recalculation was rational.

Anthropic wants power users, that's specifically their game, they just don't want those users using a harness they don't control.

> Everyone on a plan pays for the max set of tokens in that plan.

From Anthropic's perspective, everyone pays to be in bins with a given max.

And to everyone's benefit, there is a wide distribution of actual use. Most people pay for the convenience of knowing they have a max if they need it, not so they always use it.

So Anthropic does something nice, and drops the price for everyone. They kick back some of the (actual/potential) savings to their customers.

But if everyone automates the use of all their tokens Anthropic must either raise prices for everyone (which is terribly unfair for most users, who are not banging the ceiling every single time), or separate the continuous ceiling thumpers into another bin.

That's economics. Service/cost assumptions change, something has to give.

And of the two choices, they chose the one that is fair to everyone. As apposed to the one that is unfair (in different directions) to everyone.

Yes, mostly what I'm saying, but forgetting the important part:

From the email: > but these tools put an outsized strain on our systems. Capacity is a resource we manage carefully and we need to prioritize our customers using our core products

OpenClaw doesn't put an outsized strain on their systems any more than Anthropics own tools. They just happen to have more demand than they can serve and they benefit more when people to use their own tools. They just aren't saying that explicitly.

It has nothing to do with fairness or being nice.

If this was a gym subscription, it would be an equivalent of some people going to the gym, and some people sending their android to the gym every day, for the whole day, and using as much equipment as the gym policy allows.

It would be like some people sending the gym's competitor's android to the gym instead of the android the gym provides. Said gym also doesn't have enough equipment for everyone's gym appointed android despite being more expensive. Said gym doesn't want to admit this, nor does it want to raise prices on an already more expensive subscription. Said gym doesn't want competitor's android to gain marketshare. Said gym blames competitor's android for using up gym equipment despite gym's own android being capable of using as much equipment.

> using as much equipment as the gym policy allows.

which said customer paid for. And now they want to back out of it because it turns out they thought users wouldn't do that.

I say they ought to be punished by consumer competition laws - they need to uphold the terms of the subscription as understood by the customer at the time of the sign up.

> there is a wide distribution of actual use

except when people start using openclaw, and the distribution narrows (to that of a power user).

I hate companies that try to oversell capacity but hides it in the expected usage distribution. Same goes for internet bandwidth from ISP (or download limit - rarer these days, but exists).

Or airplane seats. Or electricity.

> I hate companies that try to oversell capacity but hides it in the expected usage distribution.

Except they charge you less because of the distribution. Competition for customers doesn't evaporate.

Why would you assume that to be?

They might charge you less, but they don't have to and wont if the market allows it

Companies compete by optimizing margins. Lower margins, more sales, and more customers for more forward looking sale. Higher margins, more profit per sale.

That's a "fixed" constraint, because maximizing future adjusted value is what companies do.

So they don't play little games with mass products. If they did they would be harming their own bottom line/market cap.

(For small products, careful optimization often doesn't happen, because they are not a priority.)

Note this thesis explains what is going on here. What was previously one kind of customer (wide distribution of use), is now identifiably two. The non-automated token maxers (original distribution) and automated token maxers (all maxed, and growing in number). To maintain margins Anthropic has to move the latter to a new bin.

But the customer centric view also holds. By optimizing margins, that counter intuitively incentivizes reduced pricing on lower utilized products. (Because margin optimization is a balance to optimize total value, i.e. margins are not the variable being maximized.)

The alternatives would be bad for someone. Either they under optimize their margins, or change regular customers more which is unfair. Neither of those would be a rational choice.

(Fine tuning: Well run companies don't play those games. But companies with sketchy leaders do all kinds of strange things. Primarily because they are attempting manage contradictory stories in order to optimize their personal income/wealth over the companies. But I don't see Anthropic in that category.)

The trade-off is that if you set your usage limits so that you can handle the case where everyone is saturating their limit at all times, then (1) the usage limits would be too small and (2) you're optimizing for a usage pattern that doesn't exist and (3) you're severely underprovisioning, which is worse for everyone.

Instead, you can prioritize people "earnestly" bursting to the usage limits, like the users who are actually sitting at their computer using the service over someone's server saturating the limit 24/7.

The goal is to have different tiers for manual users vs automated/programmatic tools. Not just Anthropic, this is how we design systems in general.

Well earnest here just means using Claude code directly or the Claude app. Both that just happen to support using tokens while you sleep!

Defining earnest (placeholder word btw) is the hard part of the trade-off, though.

When your least automated, most interactive users are competing for capacity with fully-automated tools, let's say, you're forced to define some sort of periphery between these groups.

OpenClaw is a self-directed, automated loop that sits on a server. It's wowing its owner by shitposting on moltbook and doing any number of crazy stories you can find online that amount to "omg I can't believe my self-directed claude loop spent all day doing this crazy thing haha."

On the other end of the spectrum is someone using Claude.app's interface.

And then in the middle, you can imagine "claude -p" inside a CI tool that was still invoked downstream of a user's action. Still quite different from the claude loop.

Claude code has /loop. Claude app has scheduled tasks. The leaked source has a proactive mode.

I'm sorry but this framing just doesnt make sense.

Even with those tools, the usage of Claude Code with all of them turned on is going to trend much lower than OpenClaw usage. Everyone that I've seen with OpenClaw will intentionally waste tokens just to make sure they hit the cap, even if they're doing useless stuff with it. And it can be going 24/7, every minute constantly, while the intended purpose with scheduled tasks is to use them at a set rate but not nearly constantly.

Definitely. They will see less usage. That's good for them because they have infra scaling issues that they don't care to admit explicitly. Their competitor will also get less telemetry (if they enable it). It's a win win.

I think's an example in my favor, though.

/loop is for polling things in a foreground Claude Code session. Their own examples include polling a build that takes a long time.

OpenClaw is much, much more than that.

I don’t really follow what you’re saying. You mention the 5 hour limit. Is your expectation that they have enough capacity so that everyone can hit their 5 hour limit all the time? Or you are proposing that’s how they limit capacity for a subscription?

Do you have an example of how this is how they have advertised or sold the plan? I don’t recall ever seeing any advertisement that their plan is simply pre paying for tokens.

This is what I've been wondering about for a while now. I have the 20x plan as well, which I thought would allow me to try some API coding - but you get zero API usage.

As you said, I would imagine where the token usage comes from is irrelevant - you are generating the same load whether you do it from claude code or some other agent. So it seems like the rules are more to do with encouraging claude code usage, rather then claude model usage.

Claude code is still getting used by these agents. They banned the mimicry awhile ago and said claude -p was fine.

OpenClaw just happens to also get telemetry, of probably higher value, out of the same tokens. It also happens to be owned by their competitor.

edit: I'm wrong OpenClaw surprisingly doesn't collect telemetry. Good for them.

If they bundled together these two radically different usage patterns, either the service would become more expensive or the limits would become a lot tighter, in both cases making Claude Code far less attractive to professional users.

OpenAI does this btw, it is why I still have that sub.

> The agent consuming those tokens is irrelevant.

This is so wrong.

The subscription is to Claude (the app, Claude code, etc) not the API.

Anthropic subsidizes Claude code because they collect a ton of super useful telemetry and logs so they can improve… Claude code.

Wanting to pay for a subscription to Claude and treat it like an API discount is like going to an all you can eat buffet and asking them to bring unlimited quantities of raw ingredients to you so you can cook at home. Ok, not a perfect analogy, but you get the idea.

> Anthropic subsidizes Claude code because they collect a ton of super useful telemetry and logs so they can improve… Claude code.

You just paraphrased my argument

Exactly.

Subscriptions are crazy subsidized.

So you can’t use OpenClaw, OpenCode, etc. because they take you outside their applications/lock in and their ability to easily monetize in the future.

OpenAI allows you to use your sub with any of these tools.

First, OpenClaw is OpenAI. [1]

Second, OpenAI is burning UNIMAGINABLE sums of money. Three days ago they raised $122 billion [2], the largest funding rounding in history. By comparison, Anthropic has emphasized a more capital efficient approach, with a ~30% burn rate. [3]

[1] https://x.com/sama/status/2023150230905159801

[2] https://openai.com/index/accelerating-the-next-phase-ai/

[3] https://www.wsj.com/tech/ai/openai-anthropic-profitability-e...

How many tokens does the $20/month buy me? I want to know what those hard token limits are but they refuse to tell me. I'm pretty sure they've reduced those limits the last week but they won't admit it. It feels like a scammy pricing model.

I agree, I think consumers appreciate transparency.

To some degree sure, is it about the number tokens you can max out?

I’m pretty happy knowing that it supports my development workflow for a week. Recent features like the Code Desktop built in browser, Cowork with Claude in Chrome and remote control matter to me way more than the number of tokens. But that’s me.

Depends on their targeted ICP also, which they are free to define. Is it those users maxing out tokens for the buck? I have the feeling there’s even better alternatives on the market right now.

> I’m pretty happy knowing that it supports my development workflow for a week

For many it doesn't. It's opaque, it changes, and they bury the news in fucking twitter. https://x.com/trq212/status/2037254607001559305

There's a lot to love about Anthropic. But man do they suck at PR.

Oh no, man fell in love with corporation

Exactly your point. Anthropic is subsidizing their own tools to keep people on their platform. What's wrong with that?

Tokens and these agents(Claude Code/cowork/claude.ai) are separate from model tokens, and they want to discount for their own product usage.

The subscription they sell is a package of these products, not tokens. They never sell token subscriptions, so why do we need to relate tokens with the subscription? Fundamentally, they never meant to sell token usage in that subscription, similar to any other SaaS company trying to sell API usage.

> What's wrong with that?

Nothing beyond fumbling the PR around it.

You’re missing something. I’m pretty sure it’s not only about the cost. Anthropic literally doesn’t have enough compute. They have to balance the load between enterprise customers and end users with subscription. If you consider they don’t have infinite compute (ie at their scale there is a limit to how much is available in a given region) and something is causing subscription users to increase usage significantly they do have to find a way to balance.

At least that’s my read. I don’t believe it is nefarious

It's not nefarious it's just bad PR cover. They definitely don't have enough compute.

OpenClaw is a mass project and doing something in the background 24/7.

I haven't even heard of claude -p before your comment.

OpenClaw is for sure not just a good cover story. Or its the cover face of the issue of automated tool workflows.

I don't think they are bothered too much about other frontends who do the same as claude code.

You are still misunderstanding.

If you max out your token limits, you are costing Anthropic more than you are paying them. They only expect a small percentage of their users to do this, but OpenClaw changed the dynamic.

Anthropic knows that they will lose more users by lowering limits than they will by blocking OpenClaw, because OpenClaw users will overwhelmingly switch to API pricing, while chatbot users will leave for competitors with higher limits.

They are a business. They hope to become profitable. This was the correct move.

What am I misunderstanding?

yes and then still subsidise subscriptions by an order of magnitude

its obvious they will tighten everything and raise prices for years to come

Look guys I use AI to help me re-write shit but for HN comments?

(Maybe I'm just being paranoid here).

-p gets penalized is not worth using it.

It’s shame they do all this sketchy stuff, I switched to Codex I have enough of their bs.

Are we now banned from using `claude -p` now?

There are multiple reasons why this makes sense for Anthropic

- The intention of subscriptions, as anywhere, is a combination of trying to promote brand loyalty, and the gym membership model of getting people to pay for oversubscribed resources that many will never use. As the parent noted, people maxxing out their allowed usage, for whatever reason, are not the most profitable customers, and in this case probably not profitable at all

- OpenClaw is now owned by a competitor, OpenAI, and Anthropic are trying to compete in this space

https://www.semafor.com/article/04/03/2026/anthropic-eyes-it...

- Anthropic are capacity constrained, having sensibly chosen to err on the side of safety (not going bankrupt), and are now trying to do the best they can to manage that.

Presumably they might be acting differently if they had capacity to spare, but even then helping a competitor to build market share in a potentially lucrative segment doesn't make strategic sense.

I do wonder about the wisdom of Anthropic promoting usage-maxxing development patterns such as running a dozen agents in parallel ... maybe not the wisest thing to do when capacity constrained! It would make more sense to promote usage at night with low priority "batch jobs" rather than encourage people to increase usage during periods of maximum demand.

Well this is what happens when everyone hires an actuary to handle their pricing and every business earns its revenue through psuedo-insurance policy subscription products.

very true.

> There's a 5 hour limit and a weekly limit. Those are hard token limits

I mean, humans sleep and do other things than work, so they likely don’t hit their weekly limits or their 5 hour limits every single 5 hour chunk :)

thanks! I never thought of using -p for using claude and gemini for one-shots and in shell scripts before. Nice.

Can’t you just use Anthropic models through bedrock?

I am happy they are banning openclaw users instead of lowering my limits to compensate for these automated agents though.

It's funny that as an avid user and developer on cc, I never used third-party harnesses, never shared my account, have one account and paid for another I just signed up for (both max 20) a more serious side project. And I'm a copius user of claude -p (spawned by their own opus [1m] model) that loops (in code written by opus) = yet I got my second account banned.

When are the honest users ever going to catch a break?

I just discovered Pi Coding Agent and found that it's lean System Prompt + a tuned CLAUDE.md brought back a lot of the intelligence that Opus seemed to lose over the last month.

Sucks to be pushed back to Claude Code with opaque system behavior and inconsistency. I bet many would rather pay more for stability than less for gambling on the model intelligence.

We use Pi at work (where we pay per token) and I’d love to use it personally too. From what I’ve read, nobody has been banned for using Pi yet… I wonder if Anthropic minds this much as long as it’s still human usage, or if they’re mostly focused on stamping out the autonomous harnesses. Unfortunately Pi is also what OpenClaw uses so it could easily get swept up in the enforcement attention.

Or maybe I’ll just get a Codex subscription instead. OpenAI has semi-officially blessed usage of third party harnesses, right?

It appears that OpenAI has blessed third party harnesses. I know they officially support OpenCode and they have this on their developer portal:

"Developers should code in the tools they prefer, whether that's Codex, OpenCode, Cline, pi, OpenClaw, or something else, and this program supports that work."

https://developers.openai.com/community/codex-for-oss

Obviously, the context is that OpenAI is telling open source developers who are using free subscriptions/tokens from the Codex for Open Source program that they can use any harness they want. But it would be strange for that to not extend to paying subscribers.

They have, but they also just announced this week that for business and enterprise plans, they’re switching from quotas for codex to token use based pricing, and I would expect that to eventually propagate to all their plans for all the same reasons.

I’d be surprised if that propagated to personal subscription plans, simply because it would put them at a huge competitive disadvantage against Anthropic, which they’ve already signaled they care about by saying they allow third-party harnesses. But I wouldn’t be surprised if they required third-party harnesses to use per-token billing, since that’d put them on par with Anthropic.

Turns out the intent to do so was announced on April 2nd as well.

regular Pi or oh-my-pi?

You can still use it with an OpenAI subscription (for now at least), and the models aren't substantially worse.

I wonder if there's a way to bring some of what Pi Coding Agent has to claude code itself.

It seems that installing claude code directly from npm shields from some of the current issues.

The entire point of AI is for it to do shit autonomously?

The whole point is that the users can have it doing shit for them instead of them having to babysit the computer.

The fact that users still have to sit there and argue with it erodes their value proposition. The proposition you can pay fewer salaries.

I would argue that „doing shit” should be done by dummy automations. AI should be used to help build that automations or step in when dummy automation breaks.

For now too many people will use AI for stuff that deterministic stupid code would be much more efficient.

They could probably offer enough tokens for that but it would be at a higher price than the sub, I think. You could still pay fewer salaries at 3k a year or per token enterprise prices or whatever.

Yes

They want you to do your shit through their own desktop apps.

It's one thing to pay $5 or $20 per month, which although it's a substantial difference, people pay that much for the convenience of having stuff ready and available - and it's a completely different thing to pay $200 per month. People don't pay that much for occasional usage and many/most people will organise themselves to use all or most of their weekly allowance when the expense is in that ballpark.

If Anthropic miscalculated the amount of tokens, or simply pushed too hard to capture market share, that is a costly mistake because people in this market are very sensitive to price hikes.

They have to be honest about what they can offer for $200. Sure, people don't max their subscriptions but when they're large they make the best of it, or they will likely cancel it. The typical subscription works well below capacity because it's cheap enough that the optionality may be worth it. $200 is not the typical subscription.

>They have to be honest about what they can offer for $200

Their expectation must have been a human using the service at a human capacity.

This is different from an automated agent orchestrating a ton of different agents at the same time doing a lot of things.

There is a difference.

You are correct, but you don't need openclaw to batch your work. People will figure out ways to use their tokens at that fixed price.

Sure there is a difference. It's like when most mobile companies wouldn't allow tethering because then people would actually use the service.

You can try to stop that, but people will price in those inconveniences. They will simply learn that the fee pays for much less than the token limit and that the company is enforcing some unwritten limits by adding extra limitations to usage.

We will see it play out.

If people are finding new ways to use AI, they should change how they bill. Banning third party harnesses is bad for a lot of reasons - it looks like they're trying to force people to use their software. Strategically it might make sense - gives them a tiny moat if their models ever slip - but it discourages the breakneck pace of innovation and the long term effect is that their customers (largely highly skilled with computers and building software) will look to decouple themselves. Claude is good but it's not so far better than anything else that they can pull shit like this and people will just deal with it.

They already have the regular subscription plans (Pro, Max) and a separate billing process for direct API usage. They could absolutely introduce another type of plan optimized toward this kind of usage or just accept that it's a dumb pipe that is being paid for and having these random arbitrary limitations is just making things more confusing and a bad plan for the future.

They already have the way that you're supposed to bill for usages like this, the API usage. The purpose of the subscription plan is strictly for the cases where you are using few enough tokens on average that it's not a money pit for them.

They don’t need to change how they bill. Your subscription is for Claude app/code. Otherwise you pay per token. It’s always been this way.

Claude Code is a subscription tier explicitly designed for agentic, automated, heavy usage. So the 'subscriptions are for human use, API is for automation' line is already blurry by their own offerings.

If the actual concern is use pattern, enforce that directly. What we have instead is metered usage + behavioral restrictions + product fragmentation across three separate offerings.

That's not a clean billing philosophy, it's layers of control stacked on top of each other with no coherent logic tying them together.

If subscriptions are for humans and API is for automation, fine. But then don't meter the human product arbitrarily and don't sell a subscription tier for automation while also restricting automation. Pick a lane.

> Claude Code is a subscription tier explicitly designed for agentic, automated, heavy usage

Except it's not. It's a desktop, web, mobile, and CLI subscription product built on top of a usage-based API with a generous token allowance bundled with it. That generous allowance comes with the restriction that those tokens can only be spent through Claude product surfaces. Why would Anthropic offer their API at a loss and subsidize the profits and growth of other businesses?

Start paying by the token if you want to use these tools. Simple as.

Even better: switch to Codex plus get better rate limits. I’m not a captive audience as much Anthropic would like to believe otherwise.

They have subscription plans for their software, and a seperate billing process for the API. There's nothing to change. 'Accepting that it's a dumb pipe' would just mean removing the Pro & Max plans as options.

Clawdbot was clearly against the Consumer Terms of Use the whole time, they’ve just started actively detecting and blocking it.

> Except when you are accessing our Services via an Anthropic API Key or where we otherwise explicitly permit it, [it is forbidden] to access the Services through automated or non-human means, whether through a bot, script, or otherwise.

The whole industry is about robots telling robots what to do, why wouldn't they have expected automation?

> They have to be honest about what they can offer for $200.

Isn't that exactly what they just did?

not really, no

being honest would be to just adjust the limits rather than adding piecewise limitations

but of course with honesty comes that people can actually gauge your product accurately and they may not want that

But the limits apply to all users. I doubt non-OpenClaw users want to pay $XX more to subsidize OpenClaw users.

it amounts to service limit obfuscation, and their market are very clued about that

nobody wants them to add fineprint every time users find effective ways to actually use the service to its advertised limits, it only benefits those who want to be milked for recurring revenue for sporadic usage while paying handsomely for the privilege

Feel free to ask for takeaway at an "all-you-can-eat" buffet. That'll show those rubes eating merely a single meal at the restaurant.

Anthropic didn't miscalculate anything. They calculated what they could charge/subsidize for humans, not automatons. Banning OpenClaw brings usage levels under control.

If you had to pay for APIs yourself for any provider then you'd know that SOTA tokens are not cheap, and Claude Code for $100 is almost a too good to be true bargain for what you can get out of it.

> People don't pay that much for occasional usage and many/most people will organise themselves to use all or most of their weekly allowance when the expense is in that ballpark.

I don't think that's accurate for professional users. Personal users, especially those for whom $200/m is a significant cost, will definitely try to get the most out of it.

I know several $200/m user (I'm on the $100 personally), and they've all had the same experience I had when first upgrading to the max package: initially you try to use it as much as you can and feel like you need to keep it busy. But that goes away after a few days and you use it when you have need. The primary point of the max tiers for my peers is to not hit limits during their work if they occasionally use it intensively because it's disrupting to have to wait for X hours to continue.

If you get a benefit from using it, and you bill at $200 an hour, and you work 160+ hours a month, the $200 monthly cost doesn't register as a significant cost and you won't make it determine your usage patterns. I'm sure that'd be different if VC money goes away and it turns out the true price would need to be closer to $5k, but at this point it's similar to your ISP for fiber costing $80 a month. You enjoy the speed for a few days, but then it becomes the new normal.

> Every single one of them oversells their capacity

This is (almost) universally true of flat rate subscriptions; but there are usage-billed ones, too (and even those often have an aspect of subsidies).

A great example of the shakeup is when dial-up went from "connect, do the thing, disconnect" to "leave the computer online all the time" - they had to change the billing model because it wasn't built for continuous connections.

That's a good analogy. Maybe soon we'll see Claude Code CDs with 700 free hours.

AOLLM!

Still very interesting timing to ban third party harnesses, given the proximity to the Claude Code leak …

Good point. I agree with that. The key point is that heavy users benefit from this model while light users are basically subsidizing them. But it's a distribution when everyone shifts toward heavy usage, prices inevitably go up. The $17/mo pro price is already set to compete with other providers. Raising it would lose customers. Other tiers are also carefully priced to match competitors. So the only move left is to prevent the whole distribution from drifting toward heavier usage. That's exatly what this ban does.

I'm pretty sure in this case it's anthropic doing the subsidizing because the api and extra usage rates are extremely expensive compared to the usage you get for the lowest subscription level. I pay $28 CAD per month and I'm pretty sure I'd burn through that in a day or two, and I'm not really a power user, I'm just using it to write code like it says on the tin. I seriously doubt there's a large portion of subscribers with low enough monthly usage that they'd save money by switching to the API.

Err, yeah, you should neither do any web scraping without respecting robots.txt, nor use ad blockers when using Google. When working with a business, never use Google Docs without paying them. Nah, that's not how the world works and at least not in the software industry.

Come on, someone on a Max account has a reason why they are paying $200. I bet many are at least often near the weekly limit, or they‘ll downgrade. If anything, OpenClaw usage is more spread out instead of ingesting whole codebases during office hours.

The Anthropic subs are likely priced at marginal cost (Amp‘s CEO recently said that in a podcast). It just doesn’t serve Anthropic to be operating as the service layer for OpenClaw.

So it's like Sliceline from Silicon Valley (the show)

> Every single one of them oversells their capacity

Indeed. And this model breaks in several cases that overlaps with the current AI business model:

- marginal cost of incremental usage is too high (Movie Pass)

- adverse selection (all you can eat monthly steak subscriptions)

- demand is synchronized (WeWork)

My impression is that at the moment the value you get out of Claude is simply incredible.

As a senior engineer, you get an assistant that never gets tired and can do quite a lot on its own. For me, it’s been an eye-opening experience. I used to have a collaborator called M that had a good general culture, but was not too smart. The calculation going into my mind every time I ask Claude for something is: how much would that cost, in terms of time and effort, to get M to do that? M was a resource that costed many thousand dollars per month, plus the time I spent correcting and directing, while Claude is actually smarter and does what it is asked with a degree of autonomy and common sense that M could never dream of.

The flipside of the coin is obvious: Anthropic will find a way to claw back - no pun intended - some of this value by raising the cost of subscription. They would be crazy not to.

value is high but what about the competitors?

is claude that good? the last time i tried claude it was sonnet 4.5. it was ok, not worth the api money clearly. but i only use api tokens for llms.

If you look at SWE, Claude models aren’t that special. Other benchmarks come up with different results.

But… anecdotally, Claude is just that good. Gemini needs a lot of hand-holding, and it will still tell you it’s done when it achieved half the work. Or say, “this test isn’t passing, I’ll just delete it”. Every now and then I get tired of it and give the same task to Sonnet 4.6; 5 minutes later I’m done. Bug fixed, UI properly working, React hooks not being conditionally rendered, theme variables used properly. It’s wonderful.

I’m not sure about large agentic work or deep thinking, but I’m mostly automating away the drudgery of dealing with React Native. I still want to do the deeper work myself, but even there Opus is usually a really good sparing partner.

Matches my experience. I am not sure why, but subjectively it feels better.

Were you using the Gemini model with the Claude Code harness? Otherwise, it is not an honest comparison.

I haven’t, but I’ve used Opus in Antigravity and it performs pretty much the same? It’s hard to tell minute differences.

Do you think Claude Code is what makes their models operate better?

And by the same token, then what would give Gemini a fair run? Because the Gemini chat app, Stitch, and the CLI are all things I’ve used and the model can’t help itself from a) saying it’s done when it isn’t; b) going off-rails; c) ignoring strict instructions after a while.

well that largely depends, lots of saas are running 90% operating profit margins

> Every single one of them oversells their capacity

That sounds like their problem, not ours

In theory yes, but the overselling does also keep the price low (at least a bit), but also boosts revenue. So when power users use the service too much, the seller will either raise prices, cut features or ban some usage patterns.

You can vote with your wallet though. So don’t throw money at them or just deal with it. Plain and simple.

Low for who exactly? You have low-users overpaying and a few openclaw users actually using what they paid for and getting banned for that... that's not really a "low price" for anyone.

If they they expect X money for Y tokens, better provide Y tokens for your X money. If they can't provide that, then change the pricing plans. That's not the users problem.

Who exactly, their intended audience. I get my money's worth having Claude Code write code. No interest in OpenClaw.

Well, yes, it is. That's why you're seeing them take proactive steps to address the problem, like this new policy.

Doesn't look like it

It's not your problem anymore once you switch from Claude :)

It's not a problem at all, you get subsidised to use it

No, people want transparency. If it was "x tokens per time interval, then you pay extra", the problem wouldn't exist.

The API offers that. Pay X per month, get Y tokens. Then you can look at all the graphs of money being deleted by OpenClaw, for transparency.

People want a free lunch. If the API was cheaper than the subscription then everyone would use the API. Instead people flock to an, apparently, unsustainable pice at a fixed monthly rate; presumably subsidized by others who don't use their full capacity every month.

So basically their move is an admission that they can't scale up their capacity accordingly to shifting demand while keeping the current pricing.

Customers have their own value calculations. If they can't use Claude for autonomous agent at reasonable price they will move to providers that are cheaper and more flexible. Autonomous agent adds way more utility than a marginally better LLM (assuming that's even true).

That is not the correct generalization. Most modern subscriptions have no capacity constraint. Usage based pricing makes more sense for a supply constrained business.

They do not advertise a capacity constraint. There is definitely one there, because it's the only way they can offer a reasonable price. Why do you think streaming services suddenly get plagued with technical issues when they host a live event? They are so chronically under-provisioned that they can only guess at the actual amount of compute they need to serve even a fraction of their subscriber base suddenly deciding to watch something at the same time. And their best guess is usually wrong, because even then — even when they know they need to deliver on a live event — they STILL under-provision their compute, or constrain their autoscaling thresholds, in an effort to save money.

Is your unlimited 5G plan actually unlimited, or does your download rate drop to dialup speeds after your crest a certain amount of bandwidth usage?

Have you ever had an ISP in a populated area? What's the reliability like?

> Every single one of them oversells their capacity.

My meal kit delivery service doesn't.

And why aren’t OpenCode and others allowed anymore?

You don’t use more tokens than with Claude Code

I run a small third-party harness myself (not OpenClaw, something much smaller). Checked my API key today after this announcement - turns out I was already on a regular API key so it doesnt affect me directly.

But the interesting thing is, my actual token usage running agents is way less than people here seem to assume. Most of the time the agent is waiting for tools, reading files, thinking. The bursts are intense but short. I probably use less tokens per hour than someone doing a long manual coding session with lots of back and forth.

The real issue for me isnt cost, its that they can just change the rules whenever. I had to drop everything today to verify my setup still works. Thats the tax of building on someone elses platform I guess.

"subsidised" is in wrong context. They charge how much they thought it would make sense then people found a way of maximizing the usage under the rules and now they change the rules. I am sure they will put out a product which is exactly OpenClaw/openclaw-like with Claude code soon, and my guess goes even to say that's the reason why they went after the naming... They totally wanted to steal the idea from the moment they saw. As they, and all other ai companies always do. They just steal and contribute nothing back.

It seems Anthropic thinks they have a much greater moat then they actually do. OpenClaw on a local model is better than any Claude offering, since it can just spin til the task is complete.

They have multiple tiers of service. The whole point of this was to allow "power users" to access more tokens. If someone upgraded to a $200/month Max subscription it's because they're a power user.

The thing is that a power user still sleeps.