Despite paying for Claude access, I don't use Fable but I find this always changing policy quite exhausting to follow. There is 2x usage, there is no 2x usage, there is 2x usage but only x hours, there is no 2x usage, there is Fable access, there is no Fable access, there is Fable access till x date, there is Fable access till x + y date. This is just exhausting. I just want to open and usage the default model and within limits. That's all I ask.
This is just a series of ploys to make more marketing about fable. They know that if they make it api only nobody is going to talk about it. And knowing that openai will have a comparable model soon means they need to stir up some hype and make people use/talk about it as much as possible. And bluffing that access will be restricted until a specific date is just marketing. Like saying buy in the next 24 hours get one free.
Yep, sounds like the usual time pressure tactics. Perhaps because no one's actually trying it.
In the mean time, their vibe coded frontend doesn't actually let me use Fable because it thinks i'm trying to hack open source software. What I'm actually doing is debugging null pointer crashes due to data corruption in the closed source software i work for these days, but no, you dirty hacker, you can't do that.
So let's recapitulate:
- they put out some marketing copy about Fable being a world ending threat for finding security vulnerabilities
- they get banned by the US government, which believes the marketing copy
- they add "safeguards" and the US government allows them to make it public again
- said safeguards make the product useless for their paying customers
I’ve consciously not used Fable because I want some usage allowance to remain for the rest of the week. The 2x billing is just too aggressive for me when I tried it (max x20 plan user currently).
> Yep, sounds like the usual time pressure tactics. Perhaps because no one's actually trying it.
For your average user, Sonnet is more than enough. Opus will be rarely used. Fable is really something professionals will use and even then with low frequency. The model is really good but the token economics don't make it so appealing to the amount and breadth of people you need to get people talking about Fable the way people talked about Chatgpt in 2022. AI fatigue is here and Anthropic and the like are fighting hopelessly against a rising tide.
Getting on and off fable this week has been quite interesting. For my personal work stream (big terraform monorepo, hundreds of states) I’ve using mostly superpowers to do heavy / quality work. But with fable, I tried just telling it what to do, and it produced roughly the same results without a big structured back and forth that I was accustomed to.
Then after using up all my fable allowance I figured let’s see if opus can actually work without superpowers, and no, it was all over the place doing weird things.
Thing is, superpowers produces meticulous specs and plans as a byproduct of its work, which is very useful for switching between work trees, stoping / resuming work by different people.
But to do that in Fable you have to spend way more tokens than it’s reasonable. You get similar quality result, but without the specs in between.
I’m not super sad that I’ll have to go back to opus though, with superpowers it was Fable but more structured. But I will miss the banter though - Fable is amazing for brainstorming big underspecced features.
I like super powers because I can be more in the loop editing and understanding the plan. Letting fable loose is genuinely impressive, and probably perfect for vibe coding, but I can’t be that far away from the plan and steps for code that matters.
As for being great for vibe coding, it’s cool but I can’t justify that kind of cost for throwaway code. At this point I’ve had good experiences using fable to review code, but I’m totally content with opus for all of my workflows still. If fable was the same price I’d switch (for projects not involving biology, at least), but I’d still use something like super powers to stay in the loop.
It's great for feature development. It queries you about you intentions, creates a spec for you to review and modify, creates a plan for it to follow, then farms out the work to subagents.
Crap on it all you want, but it makes LLM work predictable.
It does not make LLM work predictable. Try it on the same type of project twice and see. You think you're saving time but youd accomplish similar results with 2 or 3 skills.
That initially looked pretty interesting but a quick peek through issues and folks are complaining that it’s opinionated on git workflows and overriding user instructions otherwise. No thanks.
Edit: a deeper look at the issues and there are many examples of it not behaving as intended. Seems superstitious at best.
Definitely not superstitious.
It's just an opinionated workflow, you can see the steps further down in the repository.
It's just spottily enforced because it's just written to the context in markdown - it's basically one of the very first attempts from the very beginning of Claude code.
But superstitious would mean it's just in your head, essentially - but this has an effect. It does create questionairs, documents every decision and plan etc. Wherever you want that is up to you, I personally didn't... But it's also definitely very much causing an big difference in behavior from Claude Code
This. My workflow is a heavily-modified and personalised superpowers, and the docs that it produces are an asset. I tried Fable and it just ran off and did shit. Mostly that was good shit, but not all, and I would have liked to have had some input to those decisions.
I realise this is just me needing to structure my use of Fable better. But I got to a really nice place with my Opus workflow and I'm reluctant to go through that every time a new model releases.
This is an interesting comment because it makes me want to ask the question, what is the use for fable then? For me, GPT 5.4 is enough when using recursive-mode. I do appreciate GPT 5.5 Pro for some larger research, architecture, planning tasks though. I think that's what Fable is for. A very small % of total work.
>planning tasks though. I think that's what Fable is for.
this is how i've been using it, and where i've found it really excels over anything else i've tried. get fable to write a plan, and get something cheaper to follow the plan. the code fable writes isn't significantly better than the code opus writes, as long as they're both following the same plan. but a plan written by fable is much better.
What is the use of a genius who does great work but doesn’t document their results, only produces them? How much do companies enjoy having a structural dependency on someone like that?
I think it depends on your workflow. I've had a great experience with the trial. I work in research, and have set up something similar to Kaparthy's auto research. I, with Fable, have managed to get an image generation model down from 80M parameters to 10M and keep the quality of the generated images on par (similar FID). And, importantly, every change was modular, explained by Fable, reviewed by myself, and understood and documented. If not understood, I read relevant docs until I did it didn't accept it as part of the plan. So it ended up being a simple composition of existing ideas which I had previously encountered, but stacked much more rapidly than I could have.
The structure of the code is easily readable as I enforce concenventions followed by good libraries. And I can easily plug in new datasets. It's pretty good frankly.
That seems valid in today's world. Right now it's expensive, slow, and accurate. I imagine in the fairly near future it will be cheap, slow, and accurate, and that'll be a great opportunity to let it run on anything time-insensitive.
Re current use-cases: in addition to planning, there's also some tasks which Opus just can't complete but Fable can. Multiple times I've spent hours in combination w/ Opus trying to debug some particularly nasty nondeterministic issue, only to have Fable nail it in 20mins while I walk the dog.
Loosing Fable July 12th and getting usage cut by a third post July 13th is going to be rough. Looks like all the reporting about Anthropic trying their earnest to turn a profit this quarter is true. In the mean time, I will gladly switch to Codex. Their Pro plan gives you amble usage and only runs close when you are using the fast mode. And even if you run out of credits, they have given users so many credit resets this past month that you can just activate one of those after. I understand Fable is a great model, but switching to API usage will run many people thousands of dollars with the same usage from their subscription plans. Not worth it IMO.
> Looks like all the reporting about Anthropic trying their earnest to turn a profit this quarter is true.
That's not how any of this works.
Anthropic is playing the long game; they're not going make a short-sighted decision just so they announce a profit for one quarter, which doesn’t mean much because it won’t be a sustainable profit since they're going to need to spend a ton of money on training and compute over the next few months.
That's why they did a G funding round for $14 billion in February and $65 billion in May.
I suspect one reason for switching Fable to API usage for the near future is they don't have enough compute for their enterprise customers and every hobbyist on their $20/month Pro plan, 80% of which don’t need Fable 5 anyway but that won't stop them from using it as much as they can.
> which doesn’t mean much because it won’t be a sustainable profit since they're going to need to spend a ton of money on training and compute over the next few months.
Never was arguing that it would be a sustained profit, just that they are. Most likely for the reasons you just listed.
What makes you think your usage is getting cut by 50% post July 13th? It's just that you can use up to 50% of your usage on Fable right now, and that is what is going away.
I'm going to miss Fable too, I found it surprisingly tough going back to Opus when we lost Fable first time round, but paying API costs is simply out of the question for me right now.
Subscription is ~$200/month. When I tried running on tokens it was ~$500/day for the same usage pattern.
My guess is my usage is atypically high, but even if the average equivalent is $100-250/day you can get a sense of how much Anthropic is subsidizing subscriptions as a loss leader to lock market share. IMHO this is a doomed strategy since open models will get into a long tail of parity and they’ll ultimately essentially be in the business of commodity compute.
I think you need to consider how much dev mindshare you get from the loss-leader. Claude has grown at an insane rate. They brought vibe-coding mainstream.
At some point they might decide they have enough demand and inertia from enterprise to reduce the subsidy. But to say “it’s doomed” really misses the fact that it has already been immensely successful.
We can switch in a heartbeat to a competitor, what kept us to Claude was that it had better models for a while. Closing Fable off means they are squarely inferior now.
For me Opus 4.8 was a slow model with a strong habit of talking down to me in an obnoxious way that would not be possible to prompt away. GPT 5.5 is now my main driver for serious work.
In my experience, all Anthropic models tend to produce hard-to-understand word soup. They're like the kind of person who likes to use fancy words just to show you how much smarter they are.
That doesn't mean those models are stupid or just pretending. I do think that Fable is the best we can get right now. Nevertheless, I prefer OpenAI models for easier tasks where the primary output is text and explanations, not code.
I’m still rattled by all the DoD stuff in the spring. At which point I deleted my ChatGPT/OpenAI account.
But Anthropic’s games with Fable and false humility is getting a bit old. And increasingly it seems like Agent Orange is gonna implode rather than team up with someone like Sam Altman to form America’s Third Reich. Particularly post-midterms.
Nonetheless, wish there was a third option on par with them two. Maybe there is I need to investigate.
I think it's for the insane amount of data they get to collect. Most people probably don't turn off the "help improve other models" while corporations paying by the token have blanket no-train policies applied to everyone.
I also compared them against GLM Coding Plan (cause everyone was positive of the model): https://blog.kronis.dev/blog/z-ai-s-glm-5-2-is-a-great-model... and Anthropic still gives you a bit more tokens despite having some of the presumably most expensive to train and run models.
Anthropic and OpenAI are really two different businesses in one company.
One business is serving great models via the API. I think that one is indeed going to become commoditized. The other is making consumer / developer focused products like Chat GPT, Claude Code or Codex.
Anthropic is leaning hard into making Claude Code work with things like enterprise compliance policies. I wouldn't be surprised if at least one of these companies basically ends up as a SaaS fronting many different models for different capability points, some of which are open.
What makes you think the API costs are "the true price" and anywhere near their inference costs? Also what percentage of users do you think actually maxes out their subscription?
I suspect they're subsidizing a lot less than people think.
We can look at openrouter GLM 5.2 prices I think to get a rough idea about the pure inference cost (with margin). They are probably running on simmilar hardware. Although scale helps Anthropic probably.
I track the subscription value against API rates and you get between 13-20k at current rates. When Fable launched I got 32k for a short time. 500 USD per day is thus very little.
Huh, it's wild that they didn't send an email about this. I can't imagine people taking this well, especially as they don't seem to be giving notice about the end of it either.
> In the mean time, I will gladly switch to Codex.
IME Codex had a bit of a big head/ego about writing what it thinks I ought to want rather than what I actually ask for, and I've had to spend some time cleaning up its slop.
Ya, and it works. I now default to Fable-for-the-design, Opus-for-the-build mode, which is sort of cost effective, but I am hooked. Opus does not compare when it comes to architecture, designs and algorithms, esp. for gnarly problems without an obvious compromise-free solution.
Does this affect anybody? It seems they didn't reset your Fable usage so this only applies to people who didn't hit their 50%-of-plan limit with Fable and I can't imagine that's many people given how this thing eats through tokens like that's its job. On Pro I generally could not complete a single plan+execution cycle without hitting my token cap, so that 50% of weekly limit got eaten up fast and I assume that's the typical case.
I suppose it benefits people whose weekly reset is sometime between now and the 12th? Feels like vibe management, because I find this part of the promo more annoying than not.
I maximised usage and have reached the limit. I feel like I did 2 week's worth of hobby work over the last few days.
I got Fable to write multiple plans, spent a great part of the weekend reviewing them. Then with superpowers I left must of those plans executing with little intervention over the past few days.
I struggled to get Opus to just keep going without trying to convince me that it's late.
I maintain a transit website as a hobby, and I'm building a Flutter app, from scratch. The old pre-COVID one carried mental baggage.
I spent a few weekends building comprehensive plans, designs, user maps, etc with Claude. So it has enough context to make decisions and keep going.
One session lasted over a day, I imagine partly because Fable + superpowers feels slow. I have an app on my phone that I have been test running since Monday on the bus.
What really helps (not sure if Opus used to do this) is that Claude will run through the emulator on its own, verifying that the design aligns with the Figma design system we created.
This is all building on top of 15 years of existing backend and rich features, so it's not a "build me a transit platform from scratch" where AI can end up making bad decisions.
The best project I found to throw it at was cloning llama-server's web UI essentially in one shot. I'm not sure what I'll do with 5 extra days, maybe try to imagine some complex features. I'm no longer surprised at how much seems to work in these "new brain what can it do?" test, and instead think the risk is feeling like I have to take it the rest of the way once I've sunk the token cost :| https://inkcap.click
Yeah, I've been quite skeptical of LLMs, but I was wrong. The new models are capable of generating good quality code reasonably reliably. They still do some dumb things, but it's becoming more capable of one-shotting non-trivial stuff. I've no idea what software development will look like in a decade beyond absolutely nothing whatsoever what it looks like today. Even if we get sublinear progress from now on out, the current SOTA is already enough to redefine coding.
In general I expect the value of software, as a thing in and of itself, to sharply decline. With no barriers to entry, having software that just does something competently will no longer be worth much of anything. It's unclear what that will mean in the bigger picture. It'll also be interesting if this proves correct, given that software companies are largely the ones dumping so much money into this. Another probable outcome is major damage to the support-as-a-service model which again is going to directly affect many of the companies directly enabling this.
I guess the logic is that if you control the systems creating this, you control everything they're used for. But it seems equally obvious that free/local models will catch up to the SOTA today - and eventually tomorrow, so that's not a particularly realistic vision for the future.
Well, I don't know how many people it is but I certainly haven't bothered to use it except once or twice. What's the point in getting used to it as a daily driver when it won't be available in perpetuity?
But in any event, you're still right that it won't affect those of us in that camp.
It's still good for code review though. Would've been better were it not for the obnoxious cybersecurity filters that prevent the model from hardening the code.
Lots of people are trying to get the most value out of it before it goes away.
I still haven't been able to use Fable. I'm a medical physicist -- it just balks at anything I do and gives me an AUP / safeguard error. Applies even if I docstrip out all the comments in my C code!
it's incredible how volatile the product and pricing situation is right now. i struggle to find an equivalent industry that has been so all over the place and obtuse
I suspect it's the combination of wanting to capture market share (subsidised plans), being severely capacity limited in GPUs, and having bursty and absurd growth rates. There was speculation that Anthropic might be allocating fewer GPUs to training for a few days to allow people to use Fable.
The actual API pricing seems far more of a stable downward trend, if measuring by equivalent intelligence.
I kind of think they force reset last week. Which actually instead of giving me two weekly cycles of use only gave me a single week of use.
Every single thing about this is fuck users fuck your usage. All those subscriptions I bought to enjoy Fable for the time allotted? Basically gone, didn't get to use the ~4 weeks if cycles or so I had planned for, bought, anticipating. I got two. And now if I want one week more, I need to pay for a full month.
Anthropic is just the most miserable evil grinch. Everything here has totally defied everything that's been laid out for what we were told we'd get and gotten worse and worse, with less and less. Anthropic cannot general an iota of goodwill.
The most frustrating part is that everyone thought the access was ending on the seventh and used it up as much as possible only to have them say oh by the way on the 12th you can it'll go away and so you can keep using it. No one has any more fable usage tokens left? We all used it up they should've also reset people's weekly usage.
for the first time, I'm rooting for the chinese to break the american monopoly on AI. although I have gemini, anthropic and openai subscriptions, I just opened an OpenRouter account and will be using more chinese open weights going forward.
IMHE the models work much better if you go directly to the creator's API. DeepSeek V4 official API gives me 8x better caching so it was 40% cheaper in my case. Unless it bothers you China gets your data instead of giving it to the American VC-backed companies.
It also prefer to give my money to the pro-social companies giving away their open models (they cost many millions) instead of opportunists who don't give much back.
I think it does. I know this is through Microsoft, but they give you a month free of Cowork which is currently Opus 4.8 (or at least they did for us) and I doubt we'd ever go back. When I say "we" it's the enterprise organisation "we", but this also where this sort of spending won't stop anyone. I can't go into exact details, but if our first month had not been free, then our most expensive user would've hit around $1000, while our average (among users who've adopted it and actually use it) is around $100. Neither of those numbers would matter in a budget.
I've had the pleasure of setting up limits because Microsoft needed billing policies before our C-levels have even gotten it on their agenda. So I set up a sort of conservative $200 personal limit, but then setup a $1m shared limit pool that anyone can be moved into with management approval. I suspect our limits will be much higher than this once the C-levels make the decision on an actual company policy. I think we'll see these spending limits mainly used as guardrails to prevent accidental spending, but that there will not really be a ceiling, just some approval gates. Some managers are already requesting usage reports, but not to track spending, they want to see who uses too little AI.
This is the difference between enterprise and small companies and individuals. When you spend $500k a month keeping your toilets stacked with papertowels, toiletpaper, soap etc. then $1m a month on AI isn't going to raise any eyebrows.
The opacity and unpredictability of Anthropic is really starting to become more than just an annoyance. I'm glad they're extending access, but the roller coaster is really starting to cause whiplash.
If they're eventually going to add Fable to the subscription plan, I wish they'd say something about that now, or at least confirm if they don't plan to for awhile. The feeling I get is they don't want to make any announcements because they are flying by the seat of their pants and want to see what their competition does first.
Previously, access was set to expire today (July 7th)
It is theorized that OpenAI may time the release of GPT 5.6 in Codex to convert people who have lost access to Fable, so this is an interesting game theoric consequence.
Also consider that Fable launched on June 9. Many people including myself bought a 1-month Anthropic subscription just to use Fable.
By extending it to July 12, they're gonna get a second month out of a lot of such people. If it really expired today, I wasn't going to renew my month.
My theory is they saw capacity would have issues during weekday working hours due to last minute frenzy of usage, so would rather it happen on a weekend if it has to happen.
I immediately downgraded my subscription as soon as they revoked access, then upgraded it again once Fable came back. So they're getting no extra month out of me unless they keep Fable available.
Isn't the plan locked for the month though? I don't see them processing refunds for anyone who starts with Max and downgrades to Pro. Although the converse wouldn't be true: they'd be happy to upgrade a plan at any point.
It's also possible that giving access to Fable when lots of people have been taking time away from working on things (4th July, World Cup, etc) meant they didn't spend much time with it.
Anybody else find Fable almost unusable because of the safeguards? I generally can't get more than 2 minutes into a task without it auto-switching back to opus..
Kind of shooting themselves in the foot here. In the process of getting all my Fable use in, I'm also at 68% on overall week limit and 5 days left. So not only am I likely to be using OpenAI much more heavily this week, I'm going to be doing it while being slightly annoyed at Anthropic.
I am using Opis 4.8 xhigh, in OMP.sh coding agent (full agent built on Pi), with Matt Pocock Skills installed.
I don’t see a particular bump in code quality from Fable 5. In fact, it feels less reliable to me than my current setup. No sure why I am not seeing what everybody else is seeing.
Perhaps OMP/Pi (head and shoulders better than Claude Code) + Matt Pocock Skills already encode all the agentic improvements Fable has?
I had the same sentiment. In my limited testing, it didn't perform any better than Opus at all. It wasn't a particularly challenging taskset either, mostly just "add this simple feature" with plenty of context and very clearly defined scope. It worked functionally but there were much better and simpler approaches available. For the cost, I don't see how Fable can ever be worth it.
Opus 4.8 xhigh is my daily driver for everything. I'd say Fable's edge is visible when designing for a complex problem with no obvious, idiomatic solutions. It is good at greenfield designs, and good at pointing out the pros and cons of hard design choices. I now do designs with Fable, and do implementations off those designs with Opus. Pretty happy. For a while I was using Fable for everything, but burned through a lot of real money for not much value, I think for coding its slow and not at all better than Opus.
Fable may be fine but it tripped itself in the security measure and suddenly went to opus in the middle of a rework of a price index where it was improving the correction of past exchange rate…
And of course now it hides all the thinking process and just spit out the results.This make it a pain to review the proposed change and underlying reasons.
Working for a ling time with that is going to be a nightmare for anyone trying to understand what are the change for and a massive cognitive load.
I've been on Anthropic's subscription product for a few months. I pay annually.
I miss Kagi's multi-model product [1]. Anthropic's nonsense around releasing, deprecating, optimising/lobotomising is tiring, and isn't matched by the value of running different models against each other.
Fable has been fine. But its reliablity is crap. The constant downgrading is crap. This last-minute promotional windowing reeks of JCPenney pre-bankruptcy, not a trusted tool. I hate the Electron app–it's slow and ugly and shows Claude isn't trusted by its own makers with app development. I'm using 4.8 instead of dealing with the pop-ups saying my asking why basil browns is causing my account to be downgraded, and I'm still not sure if that's a nerfed 4.7.
They give me regular access to Fable, or I'm going to unsubscribe and lobby the heavens for Cohere to release a consumer AI service with a lady's name instead of something like "Claude."
I can honestly say that as a paying customer, I'm getting a bit tired of being jerked around by this company. It's on again, off again. Snip snap snip. And by the way, telling people there's a deadline so they all scramble to use their "Fable allowances" before being cut off, only to then be told "just kidding, here's 5 more days" without getting a usage reset is just another frustrating and disappointing customer experience.
Seriously, all OAI needs to do at this point is just release GPT 5.6, have it be a solid model and then not jerk it out of the hands of their customers, and they're going to eat Anthropic's lunch.
I don't think that exculpates the Trump administration for its arbitrary and capricious use of export controls, an ill-suited legal mechanism, without reasonable process or justification - thereby forcing Anthropic to scramble.
Now I'm wondering if the 84% probability of GPT-5.6 release on polymarket on july 9th is about to drop substantially (in order to release while fable is at extra cost, like everyone seems to anticipate)? If they miss thursday release, does it mean they'll release Sol on next tuesday?
I read posts like these all the time and I keep wondering when they'll begin to tighten the screws on the prosumers. I think most, if not all, $200/mo individual accounts are blasting high multiples of that amount in tokens. I mean we know that doesn't work with current inference costs, not by a longshot, so I guess this is just a way to pad their numbers like "look, we're growing our user base!" while they can still somewhat hide the "actually, we're hemorrhaging money on inference" in their accounting.
You can't drive prosumers anywhere near API prices. I would guess that the maximum you can extract from vast majority of prosumers is maybe $500/mo, and even that is a big stretch.
Once you cross that threshold, prosumers will simply fall back to using Chinese models and/or self-hosting smaller models, with more efficient and tight workflows.
Paying $100/month and using Opus 4.8 all the time to get my work done. I had a brief look at fable (when it was available the 1st time) but it burned my allowance to fast. So I keep paying the $100 for Opus and enjoy friction-less uninterrupted work.
I'm usually at 60-70% usage at the end of a 5 hour window, that's the pace where I can still think about what to delegate and what to expect and verify the results. Could probably go faster but that would have a significant impact on output quality.
I would pay vastly more, I think it’s incredible value. If I had no other choice I think I’d pay up to $1K per day. The amount of work I’m getting through is absolutely immense.
Same. I don’t know what Mickey Mouse work people are doing that 4.8 is “enough”. CRUD apps maybe?
I have some serious Bayesian statistical research programs running and Fable is on another level than 4.8. It feels like Andrew Gelman is supervising it. Even the vision model on Fable is superior to 4.8 which is great for having it digest research papers.
@anthropic, can you finally add $800-$1000 per month plan and allow us to work instead of tracking your weekly changes and dramas? I think, we (individuals, small-medium biz, first of all) did our best to help you train the model like Fable. Enterprise-level lockdown (and API costs define this) is... unfair? I mean, we all knew that you all will just use us, but it's AI, right? For people, right? Right?
The only reason this is happening -> someone (US gov?) decided that it's time to bail out those who would inevitably die within a year or two otherwise, middlemen.
There is absolutely 0 chance Anthropic or OpenAI will get a government bailout (it might still happen in this admin but it makes no sense). These companies are not like banks which are fundamentally important to the economy. Sure AI is important but Google is not going to die. Why would you save OpenAI and Anthropic when google, amazon or microsoft can just gobble them up when/if needed?
The average taxpayer gets 0 benefits from LLM. It might change in the future but for now that is true. This was exactly the reverse with banking, everyone would lose their own money if the banks just disappear tomorrow
If you look at it from the perspective of the current US administration, they see that almost all GDP growth in the past year(?) has been related to data center growth. If all of a sudden that industry is gone, you're looking at GDP stagnation or drop that looks terrible for the current party, hence the potential for a bailout.
Personally I hold the opinion that the investment into data centers would shift into something else, so no real GDP drop, but I'm not sure that's a certainty the same as 'bailout keeps the current story going'
Imo the data centers is where things start to get scary. Anthropic and OpenAI aren’t themselves indispensable, but when all of the industries downstream of them taking on more and more debt to supply projected usage in the future I could see the USG forced to bail someone out, if only to shore up their creditors.
That's not a bailout for AI labs, I meant the "bailout" for Salesforce and others. There's absolutely no place for them in the world where we have Fable+ models. For many of them. Most of them (we just didn't get this feeling yet). Someone just trying to maintain the old world order, that's all. I don't think US economy would fail if those absolutely useless giants would go down.
For coding Opus 4.8 with ultracode is near perfect and doesn't grind away as many tokens. Fable is advertised as 2x the tokens but in my experience it is closer to 5x what I burn with Opus 4.8.
I only use Fable on things Opus 4.8 would struggle too much with, and as my harness improves that’s less and less. That or they dialed up the intelligence on Opus.
Let's be real: Fable 5 on API pricing is so expensive that most solo devs won't be able to use it outside of a pro or max plan. Exceptions might be individuals with highly profitable businesses or mini side projects.
I planned to use all my allowance in 6 hours time (the time they said it would be around until) for the week and then they extend it so that I burned almost all of them in a wasteful way without having the time to review things properly.
This is the second time you posted essentially the same message 40 minutes later, are you stuck in a loop? I wonder if people are looping in addition to their AI agents.
Why don't they just continue to give you Fable access but have it use quota at 4x or 8x so they at least let people on plans continue to use it, even a little?
I'm already using it judiciously because I tried ultracode with it and it ate my 5h quota while only getting halfway through the problem.
Everyone scrambling to max their usage before the deadline may have given Anthropic some valuable data on exactly how much compute they can handle. The extension could also be a play to minimize the effect of OpenAI's next move.
Even if they grant a reset, the ball is now in OpenAI's court.
Hmm, so I planned to use all my allowance in 6 hours time (the time they said it would be around until) for the week and then they extend it so that I burned almost all of them in a wasteful way without having the time to review things properly.
I believe these labs really nailed the restrictions on independence for these models. I have seen collegues coding bs with fable 5. Genuinely, the skill issues is feel by the agent and the output is alike
Looks like we are at the end of the frontier-level models at subscription pricing. After this grace period, it will be double the cost at paid-per-token usage. I’m counting on the other models to compete at the subscription level and I need my harness to be agnostic. I need an AI harness that lets me switch LLM models dynamically depending on the task.
Perhaps. I don’t blame them for short horizons given capacity constraints, government mayhem, etc. Promising anything more than a week at a time seems fraught.
Very expensive, not sure how it fits in with many companies seemingly trying to get employees using models at about Sonnet tier pricing. I certainly would never pay API rates for it, using ccusage I am seeing myself "use" $150-$400 worth of fable per day. (https://www.theinformation.com/articles/tesla-caps-employee-...)
That's funny. Sonnet is not great in my opinion. I mean, it often amazes me with its creativity, and I do prefer its style over most bigger models from other providers.
But it also constantly just does dumb things that make me slap my forehead. Opus is a lot better but the slap ratio has not reached 0 yet. Fable seems a bit better there, more testing needed.
But yeah, VC money ran out, now we wait for Moore's Law... :)
--
Re: Sonnet-ish-pricing. GLM-5.2 is cheaper than that, and appears to be "Opus-ish" in quality. I've been having a reasonably fine time with it. My experience is that it's "very solid for small and medium tasks", haven't tested it for anything bigger though.
I use GLM 5.2 through Opencode Go, but find it to be quite slow when you get to higher contexts (80k+) through their provider so I have not been able to test too much.
Rather smart way to keep a bunch of subscriptions subscribed.
I still feel a bit salty I got so much less out of the time I thought I was buying. And I stayed up late asking Fable for what giant leaps and potentials and architectural rewrites might benefit various side projects, so I kind of got what I wanted.
But I'll probably keep one of my pro accounts, for just a bit more usage.
I half expected OpenAI to make GPT 5.6 available today, just to tempt people to switch over. Either way, I'm glad Fable is staying accessible, it's been fun.
Man, this "Fable" has been rough. Last week we get access to it again, then they reset usage after a day or two with no warning. Lots of "missed" Fable time because I was pacing it for a reset 5 days down the line. Use up all my Fable time last week, resets Sunday and I don't have enough work projects to burn up the tokens by today, so I burn them on some toy and side projects, which I wouldn't have done if I knew I was going to have it until Sunday. Now I'm at 100% and they give out more access.
It would have been WAY more useful for them to announce the extension, you know, yesterday. This is basically the worst time for them to announce it. Bunch of goobers, who thought this would be a good idea?
I'm spending more time catering to Fable and Anthropic's B.S. than solving problems with Fable. I'm increasingly convinced not getting this deeply baked into single models and going back deep learning on topics of interest is both more fun and useful. (This week: chlorophyll chemistry under heat, also the Sri Lankan civil war.)
Only inferior, talentless and lower intelligence developers need to rely on LLMs.
Fortunately, you're all going to be jobless when the AI boom fizzles out and it becomes cost prohibitive to use these products once VCs aren't subsidising it. Then us real developers will be laughing.
Despite paying for Claude access, I don't use Fable but I find this always changing policy quite exhausting to follow. There is 2x usage, there is no 2x usage, there is 2x usage but only x hours, there is no 2x usage, there is Fable access, there is no Fable access, there is Fable access till x date, there is Fable access till x + y date. This is just exhausting. I just want to open and usage the default model and within limits. That's all I ask.
This is just a series of ploys to make more marketing about fable. They know that if they make it api only nobody is going to talk about it. And knowing that openai will have a comparable model soon means they need to stir up some hype and make people use/talk about it as much as possible. And bluffing that access will be restricted until a specific date is just marketing. Like saying buy in the next 24 hours get one free.
Yep, sounds like the usual time pressure tactics. Perhaps because no one's actually trying it.
In the mean time, their vibe coded frontend doesn't actually let me use Fable because it thinks i'm trying to hack open source software. What I'm actually doing is debugging null pointer crashes due to data corruption in the closed source software i work for these days, but no, you dirty hacker, you can't do that.
So let's recapitulate:
- they put out some marketing copy about Fable being a world ending threat for finding security vulnerabilities
- they get banned by the US government, which believes the marketing copy
- they add "safeguards" and the US government allows them to make it public again
- said safeguards make the product useless for their paying customers
I’ve consciously not used Fable because I want some usage allowance to remain for the rest of the week. The 2x billing is just too aggressive for me when I tried it (max x20 plan user currently).
> Yep, sounds like the usual time pressure tactics. Perhaps because no one's actually trying it.
For your average user, Sonnet is more than enough. Opus will be rarely used. Fable is really something professionals will use and even then with low frequency. The model is really good but the token economics don't make it so appealing to the amount and breadth of people you need to get people talking about Fable the way people talked about Chatgpt in 2022. AI fatigue is here and Anthropic and the like are fighting hopelessly against a rising tide.
Spot on. OpenAI releases GPT 5.6 on Thursday.
Agreed. They've been doing this sort of thing since the beginning and it's infuriating that people eat it up. I really despise this company.
Getting on and off fable this week has been quite interesting. For my personal work stream (big terraform monorepo, hundreds of states) I’ve using mostly superpowers to do heavy / quality work. But with fable, I tried just telling it what to do, and it produced roughly the same results without a big structured back and forth that I was accustomed to.
Then after using up all my fable allowance I figured let’s see if opus can actually work without superpowers, and no, it was all over the place doing weird things.
Thing is, superpowers produces meticulous specs and plans as a byproduct of its work, which is very useful for switching between work trees, stoping / resuming work by different people.
But to do that in Fable you have to spend way more tokens than it’s reasonable. You get similar quality result, but without the specs in between.
I’m not super sad that I’ll have to go back to opus though, with superpowers it was Fable but more structured. But I will miss the banter though - Fable is amazing for brainstorming big underspecced features.
I like super powers because I can be more in the loop editing and understanding the plan. Letting fable loose is genuinely impressive, and probably perfect for vibe coding, but I can’t be that far away from the plan and steps for code that matters.
As for being great for vibe coding, it’s cool but I can’t justify that kind of cost for throwaway code. At this point I’ve had good experiences using fable to review code, but I’m totally content with opus for all of my workflows still. If fable was the same price I’d switch (for projects not involving biology, at least), but I’d still use something like super powers to stay in the loop.
(Not a Claude user) What is "superpowers"?
https://github.com/obra/superpowers
Some people just need to complicate things as a state of their natural being. What a tremendous waste of time.
That's funny, cause I'd argue superpowers barely does anything, and you need a much more complex setup to gain value.
I find it very useful. What do you use that’s better?
It's great for feature development. It queries you about you intentions, creates a spec for you to review and modify, creates a plan for it to follow, then farms out the work to subagents.
Crap on it all you want, but it makes LLM work predictable.
It does not make LLM work predictable. Try it on the same type of project twice and see. You think you're saving time but youd accomplish similar results with 2 or 3 skills.
In the three years I've been vibe coding, it's the only way I've found to consistently and reliably build complex apps. How do you do it?
Exactly. It really is such a ridiculous waste of time
That initially looked pretty interesting but a quick peek through issues and folks are complaining that it’s opinionated on git workflows and overriding user instructions otherwise. No thanks.
Edit: a deeper look at the issues and there are many examples of it not behaving as intended. Seems superstitious at best.
Definitely not superstitious. It's just an opinionated workflow, you can see the steps further down in the repository.
It's just spottily enforced because it's just written to the context in markdown - it's basically one of the very first attempts from the very beginning of Claude code.
But superstitious would mean it's just in your head, essentially - but this has an effect. It does create questionairs, documents every decision and plan etc. Wherever you want that is up to you, I personally didn't... But it's also definitely very much causing an big difference in behavior from Claude Code
Didn’t know what Superpowers was until this thread. Found this blog [0] describing how amazing Fable was at enhancing Superpowers.
[0] https://primeradiant.com/blog/2026/superpowers-6.html
This. My workflow is a heavily-modified and personalised superpowers, and the docs that it produces are an asset. I tried Fable and it just ran off and did shit. Mostly that was good shit, but not all, and I would have liked to have had some input to those decisions.
I realise this is just me needing to structure my use of Fable better. But I got to a really nice place with my Opus workflow and I'm reluctant to go through that every time a new model releases.
This is an interesting comment because it makes me want to ask the question, what is the use for fable then? For me, GPT 5.4 is enough when using recursive-mode. I do appreciate GPT 5.5 Pro for some larger research, architecture, planning tasks though. I think that's what Fable is for. A very small % of total work.
>planning tasks though. I think that's what Fable is for.
this is how i've been using it, and where i've found it really excels over anything else i've tried. get fable to write a plan, and get something cheaper to follow the plan. the code fable writes isn't significantly better than the code opus writes, as long as they're both following the same plan. but a plan written by fable is much better.
What is the use of a genius who does great work but doesn’t document their results, only produces them? How much do companies enjoy having a structural dependency on someone like that?
I think it depends on your workflow. I've had a great experience with the trial. I work in research, and have set up something similar to Kaparthy's auto research. I, with Fable, have managed to get an image generation model down from 80M parameters to 10M and keep the quality of the generated images on par (similar FID). And, importantly, every change was modular, explained by Fable, reviewed by myself, and understood and documented. If not understood, I read relevant docs until I did it didn't accept it as part of the plan. So it ended up being a simple composition of existing ideas which I had previously encountered, but stacked much more rapidly than I could have.
The structure of the code is easily readable as I enforce concenventions followed by good libraries. And I can easily plug in new datasets. It's pretty good frankly.
> A very small % of total work.
That seems valid in today's world. Right now it's expensive, slow, and accurate. I imagine in the fairly near future it will be cheap, slow, and accurate, and that'll be a great opportunity to let it run on anything time-insensitive.
Re current use-cases: in addition to planning, there's also some tasks which Opus just can't complete but Fable can. Multiple times I've spent hours in combination w/ Opus trying to debug some particularly nasty nondeterministic issue, only to have Fable nail it in 20mins while I walk the dog.
> what is the use for fable then?
Cybersecurity hardening. The one thing they don't allow the model to do.
That’s Mythos. Fable isn’t specifically for that I don’t think, just an enhanced ability and persistence of Opus.
Fable is consumer-branded Mythos.
Fable is itself a "Mythos class" model. It can do everything Mythos can do. Anthropic's classifier just doesn't allow it to do it.
Loosing Fable July 12th and getting usage cut by a third post July 13th is going to be rough. Looks like all the reporting about Anthropic trying their earnest to turn a profit this quarter is true. In the mean time, I will gladly switch to Codex. Their Pro plan gives you amble usage and only runs close when you are using the fast mode. And even if you run out of credits, they have given users so many credit resets this past month that you can just activate one of those after. I understand Fable is a great model, but switching to API usage will run many people thousands of dollars with the same usage from their subscription plans. Not worth it IMO.
EDIT: corrected usage cut amount
> Looks like all the reporting about Anthropic trying their earnest to turn a profit this quarter is true.
That's not how any of this works.
Anthropic is playing the long game; they're not going make a short-sighted decision just so they announce a profit for one quarter, which doesn’t mean much because it won’t be a sustainable profit since they're going to need to spend a ton of money on training and compute over the next few months.
That's why they did a G funding round for $14 billion in February and $65 billion in May.
I suspect one reason for switching Fable to API usage for the near future is they don't have enough compute for their enterprise customers and every hobbyist on their $20/month Pro plan, 80% of which don’t need Fable 5 anyway but that won't stop them from using it as much as they can.
> which doesn’t mean much because it won’t be a sustainable profit since they're going to need to spend a ton of money on training and compute over the next few months.
Never was arguing that it would be a sustained profit, just that they are. Most likely for the reasons you just listed.
Short term higher profit for IPO
What makes you think your usage is getting cut by 50% post July 13th? It's just that you can use up to 50% of your usage on Fable right now, and that is what is going away.
I'm going to miss Fable too, I found it surprisingly tough going back to Opus when we lost Fable first time round, but paying API costs is simply out of the question for me right now.
https://x.com/ClaudeDevs/status/2054639777685934564
I guess to be more accurate, usage wouldn't get cut 50% but rather by 33.3% post July 13th.
Subscription is ~$200/month. When I tried running on tokens it was ~$500/day for the same usage pattern.
My guess is my usage is atypically high, but even if the average equivalent is $100-250/day you can get a sense of how much Anthropic is subsidizing subscriptions as a loss leader to lock market share. IMHO this is a doomed strategy since open models will get into a long tail of parity and they’ll ultimately essentially be in the business of commodity compute.
I think you need to consider how much dev mindshare you get from the loss-leader. Claude has grown at an insane rate. They brought vibe-coding mainstream.
At some point they might decide they have enough demand and inertia from enterprise to reduce the subsidy. But to say “it’s doomed” really misses the fact that it has already been immensely successful.
We can switch in a heartbeat to a competitor, what kept us to Claude was that it had better models for a while. Closing Fable off means they are squarely inferior now.
For me Opus 4.8 was a slow model with a strong habit of talking down to me in an obnoxious way that would not be possible to prompt away. GPT 5.5 is now my main driver for serious work.
In my experience, all Anthropic models tend to produce hard-to-understand word soup. They're like the kind of person who likes to use fancy words just to show you how much smarter they are.
That doesn't mean those models are stupid or just pretending. I do think that Fable is the best we can get right now. Nevertheless, I prefer OpenAI models for easier tasks where the primary output is text and explanations, not code.
I’m still rattled by all the DoD stuff in the spring. At which point I deleted my ChatGPT/OpenAI account.
But Anthropic’s games with Fable and false humility is getting a bit old. And increasingly it seems like Agent Orange is gonna implode rather than team up with someone like Sam Altman to form America’s Third Reich. Particularly post-midterms.
Nonetheless, wish there was a third option on par with them two. Maybe there is I need to investigate.
I use them all https://github.com/yehosef/llm-tools
I think it's for the insane amount of data they get to collect. Most people probably don't turn off the "help improve other models" while corporations paying by the token have blanket no-train policies applied to everyone.
Yeah that is definitely an angle too.
Using Opus 4.8 on their Max 5x plan (108 EUR with VAT) would have cost me around 4600 EUR at API rates: https://news.ycombinator.com/item?id=48684539
I also compared them against GLM Coding Plan (cause everyone was positive of the model): https://blog.kronis.dev/blog/z-ai-s-glm-5-2-is-a-great-model... and Anthropic still gives you a bit more tokens despite having some of the presumably most expensive to train and run models.
Anthropic and OpenAI are really two different businesses in one company.
One business is serving great models via the API. I think that one is indeed going to become commoditized. The other is making consumer / developer focused products like Chat GPT, Claude Code or Codex.
Anthropic is leaning hard into making Claude Code work with things like enterprise compliance policies. I wouldn't be surprised if at least one of these companies basically ends up as a SaaS fronting many different models for different capability points, some of which are open.
What makes you think the API costs are "the true price" and anywhere near their inference costs? Also what percentage of users do you think actually maxes out their subscription?
I suspect they're subsidizing a lot less than people think.
there are many sources saying that Anthropic has 80% margin (profit) on API tokens
> I suspect they're subsidizing a lot less than people think.
Agreed.
We can look at openrouter GLM 5.2 prices I think to get a rough idea about the pure inference cost (with margin). They are probably running on simmilar hardware. Although scale helps Anthropic probably.
Well they’ve claimed operational profitability on that, and probably aren’t lying as they’re about to IPO…
I track the subscription value against API rates and you get between 13-20k at current rates. When Fable launched I got 32k for a short time. 500 USD per day is thus very little.
This is the first I'm hearing about a usage cut. Do you have a link to an announcement for this? Search hasn't shown anything.
Yeah it's right here:
https://x.com/ClaudeDevs/status/2054639777685934564
It's not phrased as a usage cut, but rather when their 50% higher limits ends.
EDIT: its 33.3% less usage after July 13th
Huh, it's wild that they didn't send an email about this. I can't imagine people taking this well, especially as they don't seem to be giving notice about the end of it either.
I started using Claude Code in the meantime with a max plan. I am very muched surprised by this.
That’s not a cut? They gave you extra usage for a set time period, it was never part of your subscription.
> In the mean time, I will gladly switch to Codex.
IME Codex had a bit of a big head/ego about writing what it thinks I ought to want rather than what I actually ask for, and I've had to spend some time cleaning up its slop.
The drug dealers are giving out a bit more of a free hit to get people hooked on the more expensive models.
Ya, and it works. I now default to Fable-for-the-design, Opus-for-the-build mode, which is sort of cost effective, but I am hooked. Opus does not compare when it comes to architecture, designs and algorithms, esp. for gnarly problems without an obvious compromise-free solution.
It's not free. It uses a lot of your quota.
$200/mo is the new ‘free’ in the area
Does this affect anybody? It seems they didn't reset your Fable usage so this only applies to people who didn't hit their 50%-of-plan limit with Fable and I can't imagine that's many people given how this thing eats through tokens like that's its job. On Pro I generally could not complete a single plan+execution cycle without hitting my token cap, so that 50% of weekly limit got eaten up fast and I assume that's the typical case.
I suppose it benefits people whose weekly reset is sometime between now and the 12th? Feels like vibe management, because I find this part of the promo more annoying than not.
I maximised usage and have reached the limit. I feel like I did 2 week's worth of hobby work over the last few days.
I got Fable to write multiple plans, spent a great part of the weekend reviewing them. Then with superpowers I left must of those plans executing with little intervention over the past few days.
I struggled to get Opus to just keep going without trying to convince me that it's late.
I'm really curious, what sort of work you were having it do such that you could execute multiple days of work with minimal supervision.
I maintain a transit website as a hobby, and I'm building a Flutter app, from scratch. The old pre-COVID one carried mental baggage.
I spent a few weekends building comprehensive plans, designs, user maps, etc with Claude. So it has enough context to make decisions and keep going.
One session lasted over a day, I imagine partly because Fable + superpowers feels slow. I have an app on my phone that I have been test running since Monday on the bus.
What really helps (not sure if Opus used to do this) is that Claude will run through the emulator on its own, verifying that the design aligns with the Figma design system we created.
This is all building on top of 15 years of existing backend and rich features, so it's not a "build me a transit platform from scratch" where AI can end up making bad decisions.
The best project I found to throw it at was cloning llama-server's web UI essentially in one shot. I'm not sure what I'll do with 5 extra days, maybe try to imagine some complex features. I'm no longer surprised at how much seems to work in these "new brain what can it do?" test, and instead think the risk is feeling like I have to take it the rest of the way once I've sunk the token cost :| https://inkcap.click
Yeah, I've been quite skeptical of LLMs, but I was wrong. The new models are capable of generating good quality code reasonably reliably. They still do some dumb things, but it's becoming more capable of one-shotting non-trivial stuff. I've no idea what software development will look like in a decade beyond absolutely nothing whatsoever what it looks like today. Even if we get sublinear progress from now on out, the current SOTA is already enough to redefine coding.
In general I expect the value of software, as a thing in and of itself, to sharply decline. With no barriers to entry, having software that just does something competently will no longer be worth much of anything. It's unclear what that will mean in the bigger picture. It'll also be interesting if this proves correct, given that software companies are largely the ones dumping so much money into this. Another probable outcome is major damage to the support-as-a-service model which again is going to directly affect many of the companies directly enabling this.
I guess the logic is that if you control the systems creating this, you control everything they're used for. But it seems equally obvious that free/local models will catch up to the SOTA today - and eventually tomorrow, so that's not a particularly realistic vision for the future.
Personally, I forgot the initial deadline was so soon and hadn't found anything to use it for yet.
I tried it but the difference to Opus didn’t seem to justify the extra cost and stress around these limits.
The Max plan at 180€/month excl VAT already comes up for budget review every time. Not sure any sort of increase will be tolerated at all.
I used up all my Fable credits over the weekend and looks like it’s completely reset as of today.
That's just going to be the weekly roll-over for you being well timed. I'm at 100% - quite annoying.
> I can't imagine that's many people
Well, I don't know how many people it is but I certainly haven't bothered to use it except once or twice. What's the point in getting used to it as a daily driver when it won't be available in perpetuity?
But in any event, you're still right that it won't affect those of us in that camp.
It's still good for code review though. Would've been better were it not for the obnoxious cybersecurity filters that prevent the model from hardening the code.
Lots of people are trying to get the most value out of it before it goes away.
I was planning on using it all until today so I burned quite a lot yesterday.
I still haven't been able to use Fable. I'm a medical physicist -- it just balks at anything I do and gives me an AUP / safeguard error. Applies even if I docstrip out all the comments in my C code!
If they pair this with a reset, then it will be good news. Otherwise we all just rushed to use our quota before the July 7th deadline for nothing.
Corollary: use your quota now because a reset seems likely.
Yes, especially in light of possible GPT-5.6, any day now. No reset = easy switch to 5.6. nobody wants to drop that usage.
I wonder if they are not just gonna wait that GPT-5.6 is released before doing any move.
I used up all my Fable credits over the weekend. Looks like mine just reset today.
it's incredible how volatile the product and pricing situation is right now. i struggle to find an equivalent industry that has been so all over the place and obtuse
I suspect it's the combination of wanting to capture market share (subsidised plans), being severely capacity limited in GPUs, and having bursty and absurd growth rates. There was speculation that Anthropic might be allocating fewer GPUs to training for a few days to allow people to use Fable.
The actual API pricing seems far more of a stable downward trend, if measuring by equivalent intelligence.
> i struggle to find an equivalent industry that has been so all over the place and obtuse
No surprises, it's fundamentally built on promises and lies
I kind of think they force reset last week. Which actually instead of giving me two weekly cycles of use only gave me a single week of use.
Every single thing about this is fuck users fuck your usage. All those subscriptions I bought to enjoy Fable for the time allotted? Basically gone, didn't get to use the ~4 weeks if cycles or so I had planned for, bought, anticipating. I got two. And now if I want one week more, I need to pay for a full month.
Anthropic is just the most miserable evil grinch. Everything here has totally defied everything that's been laid out for what we were told we'd get and gotten worse and worse, with less and less. Anthropic cannot general an iota of goodwill.
The most frustrating part is that everyone thought the access was ending on the seventh and used it up as much as possible only to have them say oh by the way on the 12th you can it'll go away and so you can keep using it. No one has any more fable usage tokens left? We all used it up they should've also reset people's weekly usage.
I really hope people cancel their subs once GPT 5.6 drops. This company deserves to sink.
How so?
for the first time, I'm rooting for the chinese to break the american monopoly on AI. although I have gemini, anthropic and openai subscriptions, I just opened an OpenRouter account and will be using more chinese open weights going forward.
It’s only a question of time. I think right now 60% of AI paper published is from China. They are already the bests at cost optimization
IMHE the models work much better if you go directly to the creator's API. DeepSeek V4 official API gives me 8x better caching so it was 40% cheaper in my case. Unless it bothers you China gets your data instead of giving it to the American VC-backed companies.
It also prefer to give my money to the pro-social companies giving away their open models (they cost many millions) instead of opportunists who don't give much back.
The drug dealer sample strategy seems to work good for them.
Does it? Maybe. But I've grown weary of it. It feels like I'm in a relationship with a mercurial partner, never quite knowing where I stand.
I think it does. I know this is through Microsoft, but they give you a month free of Cowork which is currently Opus 4.8 (or at least they did for us) and I doubt we'd ever go back. When I say "we" it's the enterprise organisation "we", but this also where this sort of spending won't stop anyone. I can't go into exact details, but if our first month had not been free, then our most expensive user would've hit around $1000, while our average (among users who've adopted it and actually use it) is around $100. Neither of those numbers would matter in a budget.
I've had the pleasure of setting up limits because Microsoft needed billing policies before our C-levels have even gotten it on their agenda. So I set up a sort of conservative $200 personal limit, but then setup a $1m shared limit pool that anyone can be moved into with management approval. I suspect our limits will be much higher than this once the C-levels make the decision on an actual company policy. I think we'll see these spending limits mainly used as guardrails to prevent accidental spending, but that there will not really be a ceiling, just some approval gates. Some managers are already requesting usage reports, but not to track spending, they want to see who uses too little AI.
This is the difference between enterprise and small companies and individuals. When you spend $500k a month keeping your toilets stacked with papertowels, toiletpaper, soap etc. then $1m a month on AI isn't going to raise any eyebrows.
The opacity and unpredictability of Anthropic is really starting to become more than just an annoyance. I'm glad they're extending access, but the roller coaster is really starting to cause whiplash.
If they're eventually going to add Fable to the subscription plan, I wish they'd say something about that now, or at least confirm if they don't plan to for awhile. The feeling I get is they don't want to make any announcements because they are flying by the seat of their pants and want to see what their competition does first.
They have an IPO to launch, so they're not going to do that.
Previously, access was set to expire today (July 7th)
It is theorized that OpenAI may time the release of GPT 5.6 in Codex to convert people who have lost access to Fable, so this is an interesting game theoric consequence.
It is, many people are hungry for GPT-5.6, considering it is supposed to be Fable class, with lower costs. Many Claude users were ready to jump ship.
Also consider that Fable launched on June 9. Many people including myself bought a 1-month Anthropic subscription just to use Fable.
By extending it to July 12, they're gonna get a second month out of a lot of such people. If it really expired today, I wasn't going to renew my month.
My theory is they saw capacity would have issues during weekday working hours due to last minute frenzy of usage, so would rather it happen on a weekend if it has to happen.
I immediately downgraded my subscription as soon as they revoked access, then upgraded it again once Fable came back. So they're getting no extra month out of me unless they keep Fable available.
Isn't the plan locked for the month though? I don't see them processing refunds for anyone who starts with Max and downgrades to Pro. Although the converse wouldn't be true: they'd be happy to upgrade a plan at any point.
It's also possible that giving access to Fable when lots of people have been taking time away from working on things (4th July, World Cup, etc) meant they didn't spend much time with it.
It was foreseen.
Anybody else find Fable almost unusable because of the safeguards? I generally can't get more than 2 minutes into a task without it auto-switching back to opus..
I'm in the biomedical field so yes, it drops to Opus 4.8 straight away due to safeguards.
Same for the dental app I'm working on.
And for the record i'm NOT vibe coding it...
Kind of shooting themselves in the foot here. In the process of getting all my Fable use in, I'm also at 68% on overall week limit and 5 days left. So not only am I likely to be using OpenAI much more heavily this week, I'm going to be doing it while being slightly annoyed at Anthropic.
Anthropic wants companies that use API billing. Their growth period through hooking individual users who bring it to the business is over.
Unless they change it so there's no data training/retention for fable, I don't even understand what company is using API fable.
I am using Opis 4.8 xhigh, in OMP.sh coding agent (full agent built on Pi), with Matt Pocock Skills installed.
I don’t see a particular bump in code quality from Fable 5. In fact, it feels less reliable to me than my current setup. No sure why I am not seeing what everybody else is seeing.
Perhaps OMP/Pi (head and shoulders better than Claude Code) + Matt Pocock Skills already encode all the agentic improvements Fable has?
I had the same sentiment. In my limited testing, it didn't perform any better than Opus at all. It wasn't a particularly challenging taskset either, mostly just "add this simple feature" with plenty of context and very clearly defined scope. It worked functionally but there were much better and simpler approaches available. For the cost, I don't see how Fable can ever be worth it.
Opus 4.8 xhigh is my daily driver for everything. I'd say Fable's edge is visible when designing for a complex problem with no obvious, idiomatic solutions. It is good at greenfield designs, and good at pointing out the pros and cons of hard design choices. I now do designs with Fable, and do implementations off those designs with Opus. Pretty happy. For a while I was using Fable for everything, but burned through a lot of real money for not much value, I think for coding its slow and not at all better than Opus.
Fable may be fine but it tripped itself in the security measure and suddenly went to opus in the middle of a rework of a price index where it was improving the correction of past exchange rate… And of course now it hides all the thinking process and just spit out the results.This make it a pain to review the proposed change and underlying reasons. Working for a ling time with that is going to be a nightmare for anyone trying to understand what are the change for and a massive cognitive load.
These comments feel like kids and their parent Anthropic.
"For summer, you're becoming a big boy so we've made the decision to increase your allowance."
You are an adult, why are you using a service that treats you like a child?
Cause all the kids are hooked on crack.
I've been on Anthropic's subscription product for a few months. I pay annually.
I miss Kagi's multi-model product [1]. Anthropic's nonsense around releasing, deprecating, optimising/lobotomising is tiring, and isn't matched by the value of running different models against each other.
Fable has been fine. But its reliablity is crap. The constant downgrading is crap. This last-minute promotional windowing reeks of JCPenney pre-bankruptcy, not a trusted tool. I hate the Electron app–it's slow and ugly and shows Claude isn't trusted by its own makers with app development. I'm using 4.8 instead of dealing with the pop-ups saying my asking why basil browns is causing my account to be downgraded, and I'm still not sure if that's a nerfed 4.7.
[1] https://help.kagi.com/kagi/plans/ultimate-plan.html
They give me regular access to Fable, or I'm going to unsubscribe and lobby the heavens for Cohere to release a consumer AI service with a lady's name instead of something like "Claude."
I can honestly say that as a paying customer, I'm getting a bit tired of being jerked around by this company. It's on again, off again. Snip snap snip. And by the way, telling people there's a deadline so they all scramble to use their "Fable allowances" before being cut off, only to then be told "just kidding, here's 5 more days" without getting a usage reset is just another frustrating and disappointing customer experience.
Seriously, all OAI needs to do at this point is just release GPT 5.6, have it be a solid model and then not jerk it out of the hands of their customers, and they're going to eat Anthropic's lunch.
Cancel your subs. Anthropic does not value or respect its customers at all.
Sure not the best user experience. But maybe they thought compute was gonna be more, or maybe fable helped them optimize under load.
I'm tired too. I'll switch to OpenAI if Sol is available on subscription models.
Agreed, but let's not forget to put at least some of the blame on the Trump administration for the export control fiasco.
Anthropic literally asked for it.
They begged publicly, multiple times to be regulated. This is before even mentioning the gross exaggeration of their model's dangerous capabilities.
I don't think that exculpates the Trump administration for its arbitrary and capricious use of export controls, an ill-suited legal mechanism, without reasonable process or justification - thereby forcing Anthropic to scramble.
Who cares, that doesn't have anything to do with the way Anthropic behaves. Not everything needs to be turned into something about Trump.
Now I'm wondering if the 84% probability of GPT-5.6 release on polymarket on july 9th is about to drop substantially (in order to release while fable is at extra cost, like everyone seems to anticipate)? If they miss thursday release, does it mean they'll release Sol on next tuesday?
I have a really hard time understanding how anyone is ready to pay $200/mo to be put on such a rollercoaster all the time.
I've used fable, it's great. But nothing beats predictability - ever.
This truly feels like some form of emotional abuse/manipulation at this point.
I pay $200/mo because my opus usage alone is >$10k/mo at api rates. Do you think I’m making a mistake?
I read posts like these all the time and I keep wondering when they'll begin to tighten the screws on the prosumers. I think most, if not all, $200/mo individual accounts are blasting high multiples of that amount in tokens. I mean we know that doesn't work with current inference costs, not by a longshot, so I guess this is just a way to pad their numbers like "look, we're growing our user base!" while they can still somewhat hide the "actually, we're hemorrhaging money on inference" in their accounting.
You can't drive prosumers anywhere near API prices. I would guess that the maximum you can extract from vast majority of prosumers is maybe $500/mo, and even that is a big stretch.
Once you cross that threshold, prosumers will simply fall back to using Chinese models and/or self-hosting smaller models, with more efficient and tight workflows.
You'd be killing your consumer line completely.
how do we know they aren't making 9800$ profit on inference from 10k$ tokens at api pricing?
Paying $100/month and using Opus 4.8 all the time to get my work done. I had a brief look at fable (when it was available the 1st time) but it burned my allowance to fast. So I keep paying the $100 for Opus and enjoy friction-less uninterrupted work.
I'm usually at 60-70% usage at the end of a 5 hour window, that's the pace where I can still think about what to delegate and what to expect and verify the results. Could probably go faster but that would have a significant impact on output quality.
I pay that much and don't even use Fable because everything I do involves biology
I feel your pain. Apparently I still do a lot of biology in my day-to-day and in my thinking, so fable jumps the tracks a lot for me.
I would pay $200/mo for Opus 4.8 already. Fable 5 is just a cherry on top (although, without it, Codex 5.5 is the better buy imo).
I would pay vastly more, I think it’s incredible value. If I had no other choice I think I’d pay up to $1K per day. The amount of work I’m getting through is absolutely immense.
Same. I don’t know what Mickey Mouse work people are doing that 4.8 is “enough”. CRUD apps maybe?
I have some serious Bayesian statistical research programs running and Fable is on another level than 4.8. It feels like Andrew Gelman is supervising it. Even the vision model on Fable is superior to 4.8 which is great for having it digest research papers.
Oh gee $2000 worth of tokens from the best available model for $200 wonder how people shell out that kind of cash
I pay $200/mo and they provide me a tool that writes all the code I want. Seems like a great deal to me.
@anthropic, can you finally add $800-$1000 per month plan and allow us to work instead of tracking your weekly changes and dramas? I think, we (individuals, small-medium biz, first of all) did our best to help you train the model like Fable. Enterprise-level lockdown (and API costs define this) is... unfair? I mean, we all knew that you all will just use us, but it's AI, right? For people, right? Right?
The only reason this is happening -> someone (US gov?) decided that it's time to bail out those who would inevitably die within a year or two otherwise, middlemen.
Seriously, just let me fucking pay to get steady service 24/7.
Use console and pay API rates. Very predictable.
No, not like that! Something that gives me unlimited usage for an insane discount!
usage credits is also an option, auto-reloads too.
American Tech companies hate their users
There is absolutely 0 chance Anthropic or OpenAI will get a government bailout (it might still happen in this admin but it makes no sense). These companies are not like banks which are fundamentally important to the economy. Sure AI is important but Google is not going to die. Why would you save OpenAI and Anthropic when google, amazon or microsoft can just gobble them up when/if needed?
The average taxpayer gets 0 benefits from LLM. It might change in the future but for now that is true. This was exactly the reverse with banking, everyone would lose their own money if the banks just disappear tomorrow
If you look at it from the perspective of the current US administration, they see that almost all GDP growth in the past year(?) has been related to data center growth. If all of a sudden that industry is gone, you're looking at GDP stagnation or drop that looks terrible for the current party, hence the potential for a bailout.
Personally I hold the opinion that the investment into data centers would shift into something else, so no real GDP drop, but I'm not sure that's a certainty the same as 'bailout keeps the current story going'
Imo the data centers is where things start to get scary. Anthropic and OpenAI aren’t themselves indispensable, but when all of the industries downstream of them taking on more and more debt to supply projected usage in the future I could see the USG forced to bail someone out, if only to shore up their creditors.
That's not a bailout for AI labs, I meant the "bailout" for Salesforce and others. There's absolutely no place for them in the world where we have Fable+ models. For many of them. Most of them (we just didn't get this feeling yet). Someone just trying to maintain the old world order, that's all. I don't think US economy would fail if those absolutely useless giants would go down.
> everyone would lose their own money if the banks just disappear tomorrow
Sure, but only because banks are legally allowed to lie to customers about how much money they have.
I was rushing to make the most of Fable today, and I'm at 67%, with limit reset on July 11th.
If would've used it more moderately if I had known in advance.
Same here. 20% usage already with just half a day usage. Wish they'd commit to keeping it on Max, even if severely limited.
Same here, I'm at 98% :(
They think they're giving you something when they're actually taking something away.
This reality show kind of bored me already. I haven’t even made an attempt to use Fable 5.
Is it so amazing? Or is it the “intermittent reward” and exclusivity driving people to use it?
It's just empty hype. It's mildly better than Opus 4.8 at best.
I think HN deserves a collective round of applause on this one: almost everyone here predicted this would happen.
For coding Opus 4.8 with ultracode is near perfect and doesn't grind away as many tokens. Fable is advertised as 2x the tokens but in my experience it is closer to 5x what I burn with Opus 4.8.
I only use Fable on things Opus 4.8 would struggle too much with, and as my harness improves that’s less and less. That or they dialed up the intelligence on Opus.
Do you mean in tokens or in $?
Let's be real: Fable 5 on API pricing is so expensive that most solo devs won't be able to use it outside of a pro or max plan. Exceptions might be individuals with highly profitable businesses or mini side projects.
I planned to use all my allowance in 6 hours time (the time they said it would be around until) for the week and then they extend it so that I burned almost all of them in a wasteful way without having the time to review things properly.
If they were going to do this, they must have known a few days in advance. Feels intentional.
It does.
This is the second time you posted essentially the same message 40 minutes later, are you stuck in a loop? I wonder if people are looping in addition to their AI agents.
I'm guessing dang might have merged some of the submissions.
I queued it up with Fable, now my limit is reached. What do you use your limits for? To check other people's comments on various threads? :)
Why don't they just continue to give you Fable access but have it use quota at 4x or 8x so they at least let people on plans continue to use it, even a little?
I'm already using it judiciously because I tried ultracode with it and it ate my 5h quota while only getting halfway through the problem.
I have a feeling they are waiting for the upcoming ChatGPT release to decide.
Everyone scrambling to max their usage before the deadline may have given Anthropic some valuable data on exactly how much compute they can handle. The extension could also be a play to minimize the effect of OpenAI's next move.
Even if they grant a reset, the ball is now in OpenAI's court.
Anyone understand what the point is of only extending it for 5 days?
I knew this was going to happen. It’s just part of Dario’s hype strategy.
Hmm, so I planned to use all my allowance in 6 hours time (the time they said it would be around until) for the week and then they extend it so that I burned almost all of them in a wasteful way without having the time to review things properly.
I believe these labs really nailed the restrictions on independence for these models. I have seen collegues coding bs with fable 5. Genuinely, the skill issues is feel by the agent and the output is alike
Still definitely not trying it if I'm gonna lose access on my subscription plan.
"First hit is free", indeed.
Looks like we are at the end of the frontier-level models at subscription pricing. After this grace period, it will be double the cost at paid-per-token usage. I’m counting on the other models to compete at the subscription level and I need my harness to be agnostic. I need an AI harness that lets me switch LLM models dynamically depending on the task.
> I need an AI harness that lets me switch LLM models dynamically depending on the task.
OpenRouter put up something about this a few days ago. Check out their Advisor and Subagent docs.
Meh... Waiting for OpenAI models without this 5 more days subscription bullshit.
It looks like Anthropic baiting people into Max subscriptions before turning the model off. No thank you.
Anthropic: reminder that DeepSeek-V4 GA version is expected to debut on July 13 as showcase for the release of the Huawei Ascend 950dt
does it matter much? i find my claude code experience always downgraded from fable to opus anyway.
On July 12th they will extend it again.
Perhaps. I don’t blame them for short horizons given capacity constraints, government mayhem, etc. Promising anything more than a week at a time seems fraught.
Anyone got any usage left to use it ?
like a fent plug on Market st
Thank you, Anthropic. This random drop in the bucket wants you to know that this was a big deal for them!
Very expensive, not sure how it fits in with many companies seemingly trying to get employees using models at about Sonnet tier pricing. I certainly would never pay API rates for it, using ccusage I am seeing myself "use" $150-$400 worth of fable per day. (https://www.theinformation.com/articles/tesla-caps-employee-...)
That's funny. Sonnet is not great in my opinion. I mean, it often amazes me with its creativity, and I do prefer its style over most bigger models from other providers.
But it also constantly just does dumb things that make me slap my forehead. Opus is a lot better but the slap ratio has not reached 0 yet. Fable seems a bit better there, more testing needed.
But yeah, VC money ran out, now we wait for Moore's Law... :)
--
Re: Sonnet-ish-pricing. GLM-5.2 is cheaper than that, and appears to be "Opus-ish" in quality. I've been having a reasonably fine time with it. My experience is that it's "very solid for small and medium tasks", haven't tested it for anything bigger though.
I use GLM 5.2 through Opencode Go, but find it to be quite slow when you get to higher contexts (80k+) through their provider so I have not been able to test too much.
Do they disclose their providers? I haven't found any info.
Rather smart way to keep a bunch of subscriptions subscribed.
I still feel a bit salty I got so much less out of the time I thought I was buying. And I stayed up late asking Fable for what giant leaps and potentials and architectural rewrites might benefit various side projects, so I kind of got what I wanted.
But I'll probably keep one of my pro accounts, for just a bit more usage.
I half expected OpenAI to make GPT 5.6 available today, just to tempt people to switch over. Either way, I'm glad Fable is staying accessible, it's been fun.
...because there's not enough token addicts and we need a proper set of whales.
https://xcancel.com/claudeai/status/2074548242386178258
Man, this "Fable" has been rough. Last week we get access to it again, then they reset usage after a day or two with no warning. Lots of "missed" Fable time because I was pacing it for a reset 5 days down the line. Use up all my Fable time last week, resets Sunday and I don't have enough work projects to burn up the tokens by today, so I burn them on some toy and side projects, which I wouldn't have done if I knew I was going to have it until Sunday. Now I'm at 100% and they give out more access.
It would have been WAY more useful for them to announce the extension, you know, yesterday. This is basically the worst time for them to announce it. Bunch of goobers, who thought this would be a good idea?
> Use up all my Fable time last week
I'm spending more time catering to Fable and Anthropic's B.S. than solving problems with Fable. I'm increasingly convinced not getting this deeply baked into single models and going back deep learning on topics of interest is both more fun and useful. (This week: chlorophyll chemistry under heat, also the Sri Lankan civil war.)
Only inferior, talentless and lower intelligence developers need to rely on LLMs.
Fortunately, you're all going to be jobless when the AI boom fizzles out and it becomes cost prohibitive to use these products once VCs aren't subsidising it. Then us real developers will be laughing.