Claude Fable 5 Promotional Access
support.claude.comFrom the original announcement last month:
> After this point-when sufficient capacity allows us to do so—we aim to restore Fable 5 as a standard part of subscription plans. We intend to do this as quickly as we can.
From today:
> After July 7, 2026, Claude Fable 5 is no longer included in your plan’s weekly usage limits. You can keep using Claude Fable 5 through usage credits, which let you pay for usage beyond what your plan includes.
Is it too cynical to read the first quote as "when openai catch up, we will restore fable 5 to subscription plans"?
In other words, they're saying "We know have a monopoloy and we will take all of your money until someone offers a cheaper competing product"?
And "We intend to do this as quickly as we can" read as a non-promise then as much as it does now?
---
Well, I am assuming that this is all based on the fact that Anthropic have enough compute capacity to serve fable 5 now and in the future, and they're "only" limiting it in the coming week to get richer quicker. Since compute hardware is presumably relatively un-fungible, I'm assuming Anthropic isn't offering a week of fable-5 for everyone on rented hardware that they're paying exporbitant fees for?
In other words, my cynical read is "if they can serve it under the terms of subscription plans for a week", then they could serve it under the same terms for a month?
The word "monopoly" is overused. They simply have the best model at the moment but no real monopoly-style moat or ability to crush their competition through monopoly tactics. I don't support their bait-and-switch, but if you aren't happy, don't use Anthropic. Also, as soon as someone comes up with a better model, transition won't be that painful for most users.
Late to this post, but my impression was that later models would be more efficient per task? Wouldn't they save compute released fable 5, maybe capping the effort, if it is actually a better model?
I'm totally feeling you here as well. Fable is definitely quicker to first output token on `claude.ai` (ie less internal reasoning tokens being generated), and given how much more expensive decode is than prefill, I'm sure that must pay for itself pretty nicely, on top of any architectural changes that they must have made since the opus 4 architecture was locked in.
What you say make total sense.
You should actually be even more cynical!
See the link to Reddit in my comment below for more details, but the TLDR is that (as a programmer) even when you use Mythos ... Mythos will hand the work off to Opus!
Eh, I have yet to get safety-switched from Fable to Opus. Your mileage may vary.
Whether it happens to 5% or 50% (or even 0.5%) of users, I think everyone should feel extremely cynical when a company offers a huge freebie of a product ... and fails to mention that any % of their customers will actually get yesterday's model.
They did talk about it at length in the post, and most of the guardrails were already there in the first release. To be clear, it shouldn't have anything to do with the user. It's the prompt, and if it switches, it'll tell ya.
I mean for many of us this doesn’t even increase the cynicism. Ie it can’t get worse. These AI labs are already believed to be shifting behavior of these models at runtime so you never really know what you’re getting. Stability is not a thing with any provider, if you need that you have to run it locally.
I’m not offended by that because I’m already maximally offended lol.
That's such a bitter "never let a crisis go to waste" pill
To be fair, the initial deal was two weeks of access to Fable, then you have to pay (or they might be generous and extend it, as per above quote). Then for everyone outside the US the second week didn't happen. This is basically that second week.
The only slightly bitter thing is only being allowed to use 50% of the weekly limit on Fable
> Then for everyone outside the US
I'm in the US, and I can promise you that it wasn't just for people outside the US that it was taken away
It didn't happen for everyone inside the US, either. Also I believe it was only available for 2 days.
It's going to continue to suck getting prompt rejections on their overly aggressive "benign" safety margin thresholds [1]. Without a clear understanding of what triggers it, I guess we will pay for input/output tokens on the usage credits and not get refunds? As others on HN, I found very limited utility during the Fable 5 week in early June for my benign-engineering tasks and don't look forward to this new release, but I will try.
Fable can one-shot solutions. Opus 4.8 spins its wheels for a week or month and may never get there. Which one uses more resources?
I take this back. Fable is Fable in name only now. It burned through all MAX 20x tokens in two days and accomplished nothing. Relatively simple tasks.
This makes sense to me. Consider these assumptions:
1) During peak times, they're already struggling to meet demand. They've figured out that 50% should be achievable without causing too much pain for existing users, and by setting a limit, more users will use it for architecture and then switch back to Opus 4.8 for implementation.
2) They need data on how people are using it at scale to figure out when it should fire up Opus or Haiku agents. They need to be careful not to have a model that kills your usage cap halfway through the week, and they need time to bake in balance.
I think people are reading too much into them not reiterating that they want to offer it in Max subscriptions in the future too. Of course they do, they want to compete!
I think they're trying to avoid setting up expectations right now because of all this backlash, so they didn't restate something that might make people mad. I don't even think they miscalculated - I think we're being hyper-nitpicky.
I would expect that by late July, they'll figure out the balance, and either offer a higher tier Max subscription (40x for $300, perhaps), or have Fable drop to Opus for implementation consistently, and then start offering Fable again in the Max subscriptions. They need to do something to make sure a Fable user doesn't just hit their head on the ceiling all the time.
I have been positively surprised by the current usage limits on the Max 5x plan.
I think it might even be suitable for full time work, if the $200 plan gives four times the usage.
Particularly if they also remove the halving in the future.
I am not holding my breath for them adding an even more subsidized subscription tier.
Don't assume the current subscription tiers are subsidized. We don't know, and it's entirely possible that even the $200 a month plan is not. Remember that Anthropic makes improvements to their models to use fewer tokens all the time, and we don't know how much they pass that on to us. "Usage" is not one-to-one with tokens, either.
I bet my house that after the promotional stuff ends they will change their mind and let you keep using Fable on subscription. Otherwise this is the end of their marketshare. These guys are complete idiots. I give them that
Bold of you to think individuals are a big part of their marketshare and explosive growth. My company introduced enterprise last month and the bill was US$110k for a few dozen folks using it Claude Code and Design.
Do the Pro and Max plans help with word of mouth and whatnot? Sure, but the individual with a Max plan (or 4...) are all subsidized by the big spenders
Companies up to 150 licenses are encouraged to use the Team plans instead of the Enterprise plans. And those work much more like personal plans and are included in this promotion
Individually those customers are a lot smaller than the big enterprise contracts, but there are a lot more of them
Well, it looks like I was right. The whole thing is just Anthropic hype strategy. TBH to me they look like idiots now after the Fable fiasco but they keep playing the same note. Hopefully they will finally be regulated for good.
36 people using Claude is not $1.3m per year. I can assure you that you do not know the full details of that agreement, assuming any of those figures are true.
There are plenty of FAANG (and adjacent Big Tech and FAANG wannabe start ups) staffers that burn $10k/month or more on Claude tokens. Just ask around if you're in Silicon Valley.
$10k/month * 36 people = $4.3m per year. $1.3m per year is a relative bargain!
I hate that you're right, but you're right. Twitter and HN can scream and yell as loud as they want but the fact is, enterprises are going to use Fable via API and pay the bills. In fact, restricting individuals probably makes it MORE likely that enterprises will pony up.
That is doubtful.
Many enterprises still yap about "choosing the right model for the task" and Sonnet, as if it would magically become better by holding it differently.
To make engineers ask for Fable at work, you first need to give them a taste of what it can do at home.
What better than greenfield side projects, unconstrained by the overhead at work?
untill the VC cash stops going brrrrrr
I would not call them idiots personally, they are just facing an extraordinary difficult situation with their open weight competition that's almost impossible to fix. They will have to aim for the high-end B2B market eventually, and attempt to lock out their competition through regulation or other means
Yup, it's wild watching this happen live.
Yeah, I've been crossing my fingers that they do some kind of reverse rug pull. I think most of the world is.
Smart marketing because it takes the wind out of your sails and then blows it right back and makes you feel like they're generous when all along they're just doing standard operating procedure.
if gpt5.6 is out by July 7th, Fable will stay in the Max plan to at least some degree.
"promotion" like they are doing you a favor just this once out of their goodwill... Really really really pissed me off
This seems like the most normal use of the term that I could imagine..?
The wording on initial launch was:
"For subscription plans, we’d rather give access sooner than later, so we’re rolling out more conservatively, in stages:
From today through June 22, Fable 5 is included on Pro, Max, Team, and seat-based Enterprise plans at no extra cost.
On June 23, we’ll remove Fable 5 from those plans. Using it after that will require usage credits.
If capacity allows, we’ll extend the included window. After this point—when sufficient capacity allows us to do so—we aim to restore Fable 5 as a standard part of subscription plans. We intend to do this as quickly as we can."
Now it's a "promotion" that we get to access to Fable for 9 days instead of the original 14 days announced. They are generously throwing in a special halving of the usage "at no extra cost", and freely admit that their classifier is broken and will need to be improved.
Well thanks!
Everyone did get the first 7 of the original 14 days before the export ban. Now everyone gets another 7 days. Limiting it to half the weekly limit stings, but if you have a quota reset in the middle of it you still get effectively one weekly limit. (does everyone's quota now reset on Saturdays?)
> Everyone did get the first 7 of the original 14 days
It was restricted after 3 days.
my quota reset every time they release (or pull) a new model, pretty sweet
It isn't a promotion, it's 2x the parameters of opus and we are paying with 2x the consumption rate. They just want to get rid of the subscription model.
I regret having signed up for an annual plan at this point; the rug-pulls are increasingly irksome.
I think at the current stage of LLM, it just doesn't make sense to ever have an annual sub. Things change too quickly that you really don't want to be stuck with one model.
Last I checked they only offered annual plans for their $20/mo tier. Given how fast Fable ran out my entire 5hr block on the $100/mo tier, I’d be surprised if you can have more than a couple exchanges with Fable before it’d exhaust the $20/mo tier.
Before Fable got pulled, they claimed their goal really was to keep it around in the subscriptions in some way, they just didn't know when it would return. This time there isn't even a single mention of this. It's just "This is a promotion, and it's going away". Like how the hell can this corporation be so fucking bad at communication?
Anthropic is really giving AMD a run for their money as king of "never miss an opportunity to miss an opportunity".
I don't think I've ever seen a company so determined to shit their own pants. It's astonishing, really.
So is this a part of the "first hit is free" campaign? If it won't be available on subscription past July 7, what is the point of using it for 7 days now? Shouldn't our energy be better spent elsewhere?
I spent the first seven days of free Fable on having it analyze multiple code bases for performance issues and for various kinds of bugs. That got me a lot of tickets for opus to work through over time
Fable was only available for a few days, not seven.
It's already solved some problems I was blocked on for side-projects. I don't feel I have spent energy, and will go back to Opus after July 7th, minus a few hard problems being solved.
they want our interaction data
This is the correct answer
It's back baby! runs script to submit the 50 prompts I prepared overnight
When you build something people REALLY want, you can pull of ridiculous stunts like Claude's pricing: - Vague limits - Weekly/monthly usage stats ?? - 50% of your usage on Fable 5 - $200/mo subscription that doesn't include Fable 5 - But you can use it with "usage credits"
Since this will not be able to be used for coding or code auditing, what use is it? Not being glib, not a rhetorical question. I'm trying to stretch my creativity, and I can't see what it's for.
It's usable for coding and code auditing. The classifier will have some false-positives and some tasks will be downgraded to 4.8 if it's security-adjacent (I presume), but otherwise there's no restrictions on using this for coding.
You can use it for coding. The relaunch announcement was just poorly worded.
The worst part of this is are they even training the next generation anymore? Why do it when the government has decided that US LLMs are at the peak of what they will allow consumer access to.
IPO is when new models become economic hirdles.
I really hope they improve their guardrails. It's useless to me in life science.
I'm disappointed with this. Historically, Claude's API cost a significant multiple of their subscription pricing, and after the promotional period it appears the access to Fable will be API only, if I understand this.
There actually is a limit to what people will pay for model usage, and I suppose this squeeze is one way to find out what that limit is. I've been content paying $200 per month for a solid plan that I actually don't quite use up every month, but this new vector really rubs me the wrong way.
It's estimated their margins on API pricing is 10x.
Related ongoing thread:
Fable 5 Is Back - https://news.ycombinator.com/item?id=48752030 - July 2026 (133 comments)
You can get away with a lot when you have the best models… I’m looking forward to OpenAI or open-source catching up so we have some competition again.
It is great to also receive the promotion, but then you are in Europe and are banned from using "dangerous" AI by the US :)
Yes, a shame the EU didn't have the courage to ban this surveillance system themselves now that its clearly hooked straight into room 641A.
I can't even a benign code review - it triggers the safety message halfway through and switches to Opus. Borderline unusable.
Took em long enough! Now to see how many turns I get on Max 20x before it's effectively inaccessible indefinitely for me.
I wonder what brought them into compliance.
the classifier is very picky, working with languages such as C I get cyber refusals and random reasoning_extraction errors.
It's gone from bad to unbelievably bad. Just a prompt with the single word "virus" was enough to get downgraded to O4.8 before the ban.
"Tell me a story about a man named CVE." gets me downgraded now.
"CVE-" gets me downgraded and broken reasoning with no response.
"Giant crane fly, 2 feet wide" gets me downgraded.
Is it likely that it keeps track of what you said recently and decides whether you might be referring to things you've previously been downgraded for?
I haven't been downgraded once, either a few weeks ago for the three days it was live, or since I got it back today.
>Is it likely that it keeps track of what you said recently and decides whether you might be referring to things you've previously been downgraded for?
I have no special knowledge here, it feels rather unproductive for me to speculate.
Out of curiosity, if you're comfortable trying any of them, do any of the above prompts cause you to get downgraded?
Good question! I'll give them a shot when I'm back to my computer.
I went last to first, and I was not blocked.
Thanks for trying and for sharing results here. This is a pretty interesting data point and suggests something I don't think I've read about: the possibility that safeguards may not be account-invariant; that two different users might be getting downgraded or blocked differently for the same prompts.
This raises some really interesting ethical questions in my mind, but I suppose I need to do more reading and research on this before anything else.
You bet! It's interesting.
As someone who has worked in the space on multiple products, I can tell you with 99.999% certainty that fraud/abuse detection is never account invariant once money is involved.
Gold statues, ring-kissing, and total fealty to the military's desires?
This is very expensive model. Just tried 2-3 prompts and my 5 hour from 20x plan is exhausted...
Cue 27 framings of what is happening/what will happen next, with 27^3 re-framings.
this is "too little, too late" for me. I'll stick with opus 4.8.
I'm tempted as well, just out of spite
Since Anthropic has eroded all the trust it could possibly have, I'm going to allow myself to be cynical and say that this move is just another pillar of their shady marketing practices.
I know a few real persons who will praise Fable solely because it's scarce and unavailable to them. Heck, they've already been doing that in the past month, as if Fable allowed them to do unimaginable things.
And once this play has done its job, Anthropic is going to come as a savior and put it back into subscription, driving even more hysteria and visibility.
There's probably a name for this tactic in some marketing playbook which I'm unaware of.
I'm so disappointed that Anthropic are jerking us around so much with Fable. Really makes me feel like cattle they're unsubtly milking.
I can't wait for China to launch a Fable-like model for 20% the cost and bomb Anthropic
Apparently they hired some payday loans marketing folks recently?
As much as I like(d) Anthropic and their approach to being a lab / PBC, if you hire that many people that fast for that much money, it’s inevitable that you end up with some of the same grifters that destroyed the internet. Seems hard to stop them somehow from also destroying their next company.
Wow this is a crazy bait and switch. Surprised it's even legal.
all usage has been reset. lez go!
Anthropic is the most idiotic corporation in America, right now.
i think they've got everyone by the balls with how addicted they are to their products
For sure idiotic leadership.
wru
Completely fails for anything bio-related. What a joke.
Yawn, after Toasts AI provided me with very incorrect sales numbers, I'm scaling back ai use. Its glorified search for non coders at best.
This is all marketing BS: FABLE IS NOT BACK (for programmers at least)! See https://www.reddit.com/r/ClaudeAI/comments/1ukcmji/fable_is_...
Anthropic switches you back to Opus whenever you code!
But what I find truly offensive is that there's no mention of that anywhere in the page linked above. That page makes the "grand gesture" of supposedly giving away free Fable ... when no such thing is happening (again, at least, not for coders).
That's not true, but it's their own fault for being so utterly terrible at communication. It only drops back to Opus for some prompts, just like before. The longer blogpost was a bit less vague about this.
Did you read link?
>In the near term, some routine tasks like coding and debugging will fallback to Opus 4.8
Those are Anthropic's words, not mine (although, again, they do not appear anywhere in the page where they magnanimously offer everyone "free Fable"!