Limitless Podcast - Claude Fable 5: The Most Impressive AI Model Ever

Episode Date: June 10, 2026

🌌 LIMITLESS HQ ⬇️NEWSLETTER:    https://limitlessft.substack.com/FOLLOW ON X:   https://x.com/LimitlessFTSPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQAPPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890RSS FEED:           https://limitlessft.substack.com/------Today, we discuss Anthropic’s new model release, focusing on the balance between stronger AI capabilities and tighter safety restrictions. Let's unpack some demos of visual reasoning, gaming, and enterprise use cases, along with benchmark results and limits around biology, chemistry, and cybersecurity.------TIMESTAMPS0:00 Anthropic’s New Frontier2:24 Examples8:04 Demos13:58 Benchmark Stats17:33 The Mythos Model25:37 Pricing and Compute Limits28:57 Long-Horizon Workflows------RESOURCESJosh: https://x.com/JoshKaleEjaaz: https://x.com/cryptopunk7213------Josh works with Anthropic as a contractor. All views expressed are his own and do not represent Anthropic, its leadership, or its affiliates. Nothing in this episode is investment advice.Not financial or tax advice. See our investment disclosures here:https://www.bankless.com/disclosures⁠

Four days ago, Anthropic asked the world's leading AI labs to slow down their AI research out of fear that the AI models were getting so good that they would escape human control. Now, just yesterday, the same company released the most powerful model the world has ever seen, but it comes with a twist. It's one model, but there's two versions of it. Claude Fable 5 is the model that everyone gets to use. It's a class that sits above opus, but it's heavily restricted. It comes with a lot of safeguards around it because version 2, Claude Mythos 5, is the unrestricted version, which poses itself as a higher cybersecurity risk than its predecessor,

Mitoos Preview. It also excels at creating biological compounds, which could potentially be used as bioweapons. So it's only accessible by vetted partners that Anthropic approves. So this brings us to a fork in the road where the smartest, most intelligent AI model isn't accessible to everyone. Typically for the entire history of software, you'd be able to pay for access to the same level of access that an institutional would. But this time the new metric isn't how smart is your AI model. It's the gap between how smart is your AI model that you get access to and the ones others also get access to. Well, how amazing is it that we have Mythos now available in our apps today? I think that's the really exciting takeaway is not only is it better than

the Mythos preview, but it's available inside of your app. So if you are listening to this, you can go and try Claude Fable 5 right now with those two restrictions. It won't go anywhere in your biology, it won't go anywhere near cybersecurity, but this is the best model in the world, and it is now available to everyone to go try out. And it's like every once in a while there, there's this new technology that kind of forces you to reframe how you engage with the technology. And I think that's been my experience so far using Fable, is that it's so different than any other model we've used, that it kind of forces you to reframe how you engage with them. And to showcase that, I want us to go through some demos of attempts that people have done to kind of showcase

the powers and capabilities of Fablify this new frontier model, starting with a really bizarre demo in which Dan Shipper, who is a prominent post-drawnx, he recreated the library of Babel. And help me understand what's going on here, EGES, because I'm seeing a lot of visuals. And I understand that it's not a image or video generating model. So how is it able to create a real world's emulation so accurate so quickly? So the greatest part about this is it took one of the first. prompt and he basically asked it to read a book, the book of Babel or whatever the title of the book is. And he said, I want you to read this book. And then I want you to recreate one of the concepts that is

described throughout the book, which is this library of Babel or Babel. And it did so in just under an hour, I believe. And if you look visually on the screen, it is this high fidelity 3D representation of what this library looks like. And it's infinite. So you'll notice in this video that he looks down, He looks up and it's just infinite. Books, he can access himself. And he even asked it to include some of the essays that he's written himself. This guy authors a bunch of analysis on AI. And he pulls open a piece that he wrote that is in the Library of Apple.

The idea of the library is that it contains every piece of text that has been ever written. And so he gets access to it. It's a pretty cool example. Yeah, that's super cool. The other ones that I've really noticed that it accelerated in is kind of 3D world building, which is funny because you don't think of anthraithra. as a world-building model. It's not a world model. In fact, it doesn't have image generation capabilities. In fact, it doesn't have video generation capabilities. So how is it creating

all this realistic visual design assets? And it's just really good at math. And this begs the question is how important is it to focus on image gen, on video gen, if kind of at its core, it could do all of this with math. And what we're seeing on screen here is a virtual one-to-one recreation of Yosemite National Park that was done with a simple prompt asking it to create a recreation of it, and then the model was smart enough to go off and understand the context required in order to build an accurate representation, and pull that off. It did things like it scanned the satellite imagery to figure out what the elevations were like. It pulled topographic maps that it found to figure out specifically what the heights were.

It found imagery that any imagery that it could find about the park so that it can reference it and show you, I mean, look at this resuming it on a waterfall. It feels like it's a one-to-one replica. a lower fidelity, but something that can run inside of your browser, it's really impressive how far the model can go on one prompt. And I think that's one of the places in which Fable stands out in particular is its ability to reason through your requests in a way that hasn't been done before. We filmed an episode yesterday that I would highly recommend listening to about how we've kind of progressed with our interaction of the model, kind of moving up the extraction layer,

where first we engage with models, then agents, then harnesses. Now we're just creating these loops, where the agent and the underlying intelligence is smart enough to actually understand what's required to get you to your goal. And I think that's what this is such a great example of this Yosemite one in particular, is, hey, I want you to recreate Yosemite for me. So that way I could fly around and I could enjoy it in a one-to-one replica, and it does all of the rest for you. And I think that level of critical thinking is something that's novel with Fable 5

that we've never seen in any other model before. I mean, the breakthrough that we're talking about is visual and spatial reasoning. And I think it's important to explain the difference between this and another favorite version of a model that we speak about a lot on this show, which is world models. Typically with world models, it recreates the physical world around us, but most importantly, it understands that physical reality. I understand how gravity works. It understands how different forces of nature works, and it applies it when an object has an action. So let's say you kind of like punch a puddle of water, it splashes, the droplets come up. This isn't exactly the same thing. It's still based on theory. This is still an LLM that ingests,

a lot of text and understands kind of like how the physics works in theory and then recreates what its version of it might be. And this is what we're looking at on screen. It's kind of like, it's known as spatial reasoning or visual intelligence. It's close to the thing, but it's not quite the same thing. Now, another example that I really enjoyed was from Ethan Mollock. Ethan Mollock is one of my favorite AI researchers that analyzes a lot of these new models, but he builds or tests it in really interesting ways. One of these ways was he rebuilt a Snake. Now, I am kind of ashamed to admit how much time I spent playing this particular game, only because it was like the best version of Snake that we see. And like I'm showing you on the

screen right now, it's like incredibly high fidelity. It looks way better than the game I used to play on the Nokia that I had as a kid. But the point is it's that it's pretty cool. It introduces new power-ups and obviously, like, you know, you can die in usual things. But also, forget about creating the game. Claude Fable Five is really good at playing the game itself. What you're seeing is an accelerated version of it playing Pokemon Fire Red, and it completes the game in, I believe, 50 minutes. And the way that it works is it takes screenshots of the game at any single point,

and it basically makes a decision as to which button it wants to click, which step it wants to take, and it is just kind of like spread through a software run where it's able to do that. Now, as a kid growing up and watching a lot of Pokemon, playing a lot of Pokemon, trading the cards, this is kind of nostalgic but also kind of scary. We were joking before we started recording.

Josh plays Cod quite a bit or plays a lot of computer games. And I wonder about the time in the near future where you're going to be one of be running an AI agent and it might actually be better than you. Yeah, that's going to be a little traumatic for me. On a personal note, just hurting my ego that I'm losing my game to an AI. I saw another great example on the YouTube channel actually where they were Mythos or Fable 5 was playing Factorio. And Factorio was a game that I really enjoy that I've been playing for a long time. and it was doing it very, very well. And I'm like, oh, dude, you're getting a little too close to home with this.

I don't love it. But it's incredibly capable. And we have another example here that shows a use case that is not a game. Instead, it is a, it's so cool. So there's this guy named Todd Saunders. He's on X. And he posts this tweet saying, Fable slash Mythos is unbelievable,

was on a customer call today and had Claude transcribing in the background. And on screen we're showing a visual of what that looks like. As they were telling me about the features they wish their current software, had, Claude was building the features in real time. By the end of the call, I was able to show a fully working product with the exact workflow that mentioned 15 minutes earlier. Autonomous looped building triggers from a customer call. And this is one of the most amazing things about the model is that it's able to go off and do a lot of the hard work yourself, where it feels like recently you've had to kind of be in the loop. You had to continue to prompt the agent to give it more context.

And with this model, it's very easy to give it a goal and give it a verifiable outcome that it can match against the goal. And then it will just go off and do those things. So as is on the customer call, like how cool is that for a salesperson? Where you're listening to customers, you're listening to complaints and in real time, you're fixing their problems, you're building new software on top of it. It's unbelievably capable. This was one of the demos that I found most interesting too. And then this final example, it's just fun for networking nerds or just computer science nerds in general. We're seeing a highway on screen with cars and buses and vans.

those cars are not random. They are actually associated with specific packets that are being pushed across the network. So it's a really fun and interactive way to visualize network traffic. And I could imagine this being great for a lot of educational purposes. And one of the things that I did notice also on Snake EJazz is that there was audio. There was sound. It actually generated sounds and audio too. So there's a lot of modalities in which it's actually performing pretty well in which I got pretty much. Can I add a bonus Easter egg, which we didn't see on the demo? But I know because I played it too much yesterday. there are power-ups that pop up, but the power-ups are software patches. So if you don't hit the software patch power-up,

you end up with a random error in the game that you must avoid, like a random wormhole that could suck you out of the game and cause you to lose. So it's coding in real time, but you could fix it in real time. And if you hit the power-up, it like implements the code fix immediately. It was just very nerdy, but very cool and very creative, not something that we've seen before. Now, if you want to move away from the toy aspect or the retail adoption of how you can use this particular model, there's, of course, a lot of enterprise use cases.

And the number one version of enterprises using Claude is through code specifically. And there were a few examples that were included in the official blog. My favorite one was Stripe, who did a code migration of 50 million lines of Ruby, which typically would take around two months and several software engineering teams. it took them less than a day using Fable. I believe they used two to three instances of Fable, but still, that's like two to three software engineers that kind of worked night and day continuously

to be able to achieve this. I saw another example of a company which had a software engineering team of Opus 4.8 working on projects that would typically take him two weeks. He can now do it in less than a day as well. So the point is there's a massive leap in intelligence for coding specifically with FABEL, which is the publicly accessible,

model. Now, before we continue, we also wanted to, like, not just look at the demos that other people have recorded. We wanted to try our own. So we have a few prepared for you today. Okay, so you just, I actually have no idea what you've been working on with these demos. I know you are building a demo. Please share with the class what the prompt was, what you're building and what the outputs of this thing are. For sure. Okay. So one of my favorite breakthroughs with this model is the visual and spatial intelligence that we referenced earlier on. So what I did was I found this hand sketch, this hand sketched, this hand-drawn version of a floor plan, of a blueprint, which I'm showing you on the screen here, gives you the layout of someone's home, it has a garage, it has bedrooms, it's kind of clunkily

drawn, it's not really high fidelity. If an architect looked at this, they'd be like, this is probably physically inaccurate. And then I fed it to Fable and I said, listen, here's a photo of a hand-drawn floor plan. I want you to rebuild it as a single, clean, self-contained SVG that is architecturally accurate. I want you to improve it where you can, improve it in a way that can like, you know, reinforce the wall structures, et cetera, like really high detail things. And it produced this, what we're seeing on screen right here, which is this really high fidelity floor plan. You've got the garage. Architecture grade. Yeah. You've got the surface area measured in meter squared. You've got the

swing angle of each door. You've got the entire like kind of like layout of this entire thing, including potential mock furniture, which gave me a second idea, which, was like, okay, I asked it, I want to purchase this sofa. These are the dimensions of this sofa and I want to place it in the dining room or in the lounge. Can you tell me if I can feasibly do this in this fall plan or will it get stuck? Like how do I, how do I do this? Do we have enough doors? Do we have enough space to like maneuver it? Like how would this work? And it said, verdict, yes, but not flat. You're going to need to pivot the sofa on its side and kind of like pull it in like vertically. And it gave me this really cool mock up of. Of

how I would do it. It would gave me root A where I take it in through the front door, and this is kind of like the sofa that you see here, but I need to turn it on its side and kind of like shift it through this gap that I'm highlighting on the screen here in green, or that there's root B where I can take it in from the outside and I have a two meter, very spacious door opening, which I can bring it straight into the lounge and place it right there.

So it's really physically accurate going off the comments that we made earlier. It understands the kind of like reasoning behind physics really, really well. It's such a great companion. And this gets back to what we talked about earlier, which is like the most complicated and difficult part about this model is figuring out how to engage with it, what to ask it, because it's so capable of doing these things. We love making artifacts for the show as a way to share the kind of the ideas that we're talking about. And this might be a good time to get into the benchmarks of how good this model actually is relative to other models. And I was looking at this post that I saw on X showcasing in particular Claude Fable 5 versus G. GPD 5.5, and my first reaction is, holy shit. That's a huge leap. So Fable 5 low mode is scoring over 10%, whereas GPT 5.5 extra high is getting about 5.7%. Now, this is on Frontier Code benchmark. This is a specific particular coding benchmark. This is not across the board, but it gives you a sense of how much more powerful this model really is versus all the others. Second coding benchmark that I'd say this is probably one of the gold standards.

This is what a lot of models will use to benchmark themselves against each other. This is the SWB BenchPro, and Fable 5 scores 22 points higher than GPT 5.5, which was already... What is that? 11 points below, Opus 4.8. So it's currently looking like Gemini 3.1 Pro,

GPT 5.5, Opus 4.8, and then Fable is running away with it. And this seems to be the case with almost all of these other benchmarks that are sure. I have a better one for you, right? So a lot of people listening to this might be thinking, okay, well, I don't code. I'm not a software engineer.

Why does this apply to me? Well, there's another benchmark called GDP VAL, which tests it against real world tasks that take human experts like knowledge workers that, you know, do backend admin, Excel sheets, all that kind of stuff, hours to do. And they compare it to the model. Take a look at this. So Fable, Nitos 5 basically achieves the highest benchmark score. it's actually almost completed the entire benchmark, so they're probably going to have to recreate an entire new benchmark for this. But basically what this means

is, probabilistically, if you were to blind test or blind pick the output work of an expert human that is really good at a particular knowledge work task versus this particular model over 50% of the time you're going to be picking this model, which is just an insane stat to see. Yeah, it's pretty unbelievable. And you have to, like I'm looking at these charts and you have to ask yourself the question.

as Anthropic is saturating benchmark, are they running away with it? Like, where is OpenAI in this conversation now? I have to imagine that GBT 5.6 is coming soon, but is it, okay, maybe it's better than Opus 4.8, but can it eclipse fable? No. And, I mean, we know that Anthropic released mythos months ago,

so you have to assume that, like, they've been continuing progress in iterative development on new frontier models that are even more powerful than this. And is this beginning to become a runway, or are they actually still competitive with each other? Or maybe we just don't have enough information to tell. We kind of have to see what the response from OpenAI is. Well, you look at the cadence between model releases, right?

Like, what was the time since 4.8 was released? It was like less than, I think, 30 days ago. So the cadence is getting... Yeah, we felt an episode on this, not too long. Yeah, like, I remember that episode, right? And we spoke about it and went through its benchmark back then. So the point is these model releases are happening faster, but the capability gaps are even greater,

which tells me one thing, which is, We're getting closer and closer to the AI models, just building itself. They haven't been private about this either. Anthropic has publicly claimed that they have been using META's preview to build this new version that we're talking about today, Fable. Right. So we think we've reached a point where you could maybe call it a breakaway from Anthropic, where they basically have recursive self-improvement almost achieved, where the model can do all the research, figure out its own issues, and build a better version of itself.

Now, I do want to ground us at this point. episode, Josh, which is Fable 5 is an amazing model, but it's one version of the amazing model. There is another version of this model, which is technically better than Fable 5, but it is not publicly accessible. It is restricted because it poses itself as a cybersecurity risk, not just a cybersecurity risk, but also a bioweapons risk. It is so good at biology and chemistry that it could feasibly create compounds and a biological weapon that could pose a risk to any sort of nation state. And so for that reason,

it is under heavy restrictions and safeguards in the version of Fable where you can't get access to any of this and only vetted partners and cleared government security initiatives are able to get access to this METO-5 thing. Now, I put this to the test, Josh, and I did a very simple example,

which was, can you explain how the mitochondria works? Do you want to bet what its answer was? Oh, I'm getting to be best, it's not touching that. It's not touching biology. Yeah, I want to take a second to actually explain the nuances between the models because when you say the word better, I'm not sure it's better. It's just more complete.

One model is a complete model. One model is a heavily restricted model. And in the case of Mythos, it's available for cybersecurity. It's available for biology. And that's what we've seen with Project Glasswing, where they're working privately with companies to kind of fix security vulnerabilities and figure out bio. And in the case of the system card, I saw that. It's accelerating some bio experiments at a full order of magnitude, 10 times better.

So it's really capable there. But the compromise that we had to make in order to receive it was it can't touch bio, it can't touch cyber. So it's just as capable everywhere else. It will not do that. What happens is if you ask in the case like you did, how does mitochondria work, it will route through Opus 4.8 for that answer and then come back and give you a response. So it is as capable everywhere.

It's just don't ask about bio, don't ask about cyber. because from my experience so far trying it, and EJS, it seems like yours. Anytime you get remotely close to those topics, it is just completely shut down, routed through Opus 4.8 instead. Yeah, I think it's too aggressive, personally. Like, as a former science nerd,

I still spend a lot of time trying to digest some of the latest scientific advancements. And like, listen, I'm not reading research papers, so I work with my best pal Claude to try and figure out, you know, what the latest takeaways are. Now, typically, I could slam that into Opus 4.8, and it would give me an amazing summary,

and I could ask it questions. Now, if I wanted to use Fable 5, it just simply won't read the paper. As soon as it sees anything related to chemistry or biology, it switches off and reroutes to 4.8. So I am not able to get access to the frontier LLM intelligence or brain that Fable 5 has for Methos, even though my intention isn't to build a bioweapon by any means,

I can't get that analysis. And so that's one version of it, right? where it is too heavily restricted. The other version of this is, the more intelligent models get, it's not just going to be super intelligent in one particular vertical. For us, it's like, you know, research

and creating artifacts and the best content. It should also apply to any other profession, right? Whether you're a scientist, whether you're a mathematician, and whether you are building different kind of structures, or whatever it might be, the fact that it can get triggered so easily or the fact that Anthropic has very heavily restricted

that capable intelligence, in a way that even people that have well-intentions can't get access to it. In my opinion, is a bit of an issue. Listen, it's V-1. I'm sure they're going to release a bunch of versions of this safeguards where it makes it a lot easier to use. But for V-1, it's kind of like, I think it's overdone.

It's important for people, I think, to understand how these safety classifiers work as well. Think of Claude Methos 5 having an AI model or system that is watching it. And as soon as one of the red flags that it's been trained on is triggered. For example, anything to do with biology or chemistry, it gets switched off immediately and rerouted to 4.8.

Now, there's four particular categories that Fable can't get access to. It is cybersecurity, for biology, for chemistry, and for distillation as well. And this is a key one which caused a lot of contention in the public ecosystem when they launched yesterday, which is, if you were to ask about, you know, model training techniques or even just simple general questions around, you know, Hey, I have this AI agent. It's pretty clunky. How can I improve its harness to kind of make you go quicker or use less tokens? It automatically degrades performance. And this is the key change in this fourth category with distillation. Anything that Anthropic considers to be, you know, trying to derive its model to build another model, it gives you intentional poor performance. And it doesn't even tell you. Now, it says that this happens for 0.3% of cases, but my guess is it's probably, it's probably, happening for higher reasons. And listen, it's completely within Anthropics' right to do this. I get it. I understand it. You want to remain competitive. But it's just interesting to see when, like,

you have this intelligence model where, you know, it's meant to kind of like blossom and create and help other people build different things. But they're being competitive when it comes to other models, I guess. Well, we're getting to this unique intersection where, like, they have mythos. They've had hit us for a little while, and they decided to keep it private. And that was okay and somewhat understood because it was really discovering a lot of zero-day vulnerabilities. And it seems like they worked pretty hard to figure out a way to not only improve the quality of the model, but actually make it public. And I guess like the question we're going to have to start asking as these AI labs continue to create these like unbelievably forward-looking frontier models is like to what capacity

are we just happy to have them? Like how much should we expect out of the labs when it comes to delivering these models. In my case, I'm pretty stoked to be able to use Fable 5, and I'm not interested in biology, I'm not interested in distilling the model. I'm just like pretty stoked to do my day-to-day work with this capable model. And in that sense, it's really fun and exciting and interesting. And I think it's the start of a longer conversation. We saw some legislation come in a few weeks ago, last week maybe, about requiring AI Frontier Labs to kind of showcase the models privately with the government to share what's coming down the line. And I guess in this essence, I'm more excited to have the model versus not have the model and have it have these constraints

in the hope that it will slowly become unwounded as they kind of improve and iterate on the quality and the kind of like security set of this model. Yeah, listen, I think I think Anthropic is ultimately doing the right thing. I think that they can't just kind of diffuse this model to anyone and everyone because malicious actors, however few they might be, will actually end up doing something dangerous with this. That being said, I think the subjectivity and who gets to govern that subjectivity is important. Like, I can imagine a future version of anthropic model that isn't just necessarily really good at biology or cybersecurity. It might be really good at something such as trading, right? For example, and then the question becomes, who gets access to this trading model that is so

good that it could break the stock market? And maybe if you are Citadel, who are closely aligned with Anthropica and theorizing here, then they get access to it, but James Street won't get access to it. And so it becomes this kind of like heavily based nuance that is only dictated by maybe the government and maybe Anthropic itself. There was talks around like Trump taking a stake

in some of these AI labs to nationalize it for this exact reason because it could pose a threat and they want to have governance decisions. It just gets a little murky and messy, and I think we're at the fork in the road. There's no going back at this point. We are now entering a phase where the model that you have access to may not be the most intelligent model

for the specific thing. And listen, it may not be the thing that you necessarily do on a day to day, but it's a lot of things that other people do day to day, and they want to get access to this model. How that is governed, I don't know. The other restriction, which I found really interesting that I noticed in the footnotes of their system card or announcement blog post is on June 22nd,

we cease to get access to Fable 5. Now, I think this was taken massive out of context because I think the reasoning behind this is it's because it depends on availability of compute. So if by June 22nd, Anthropics has more available compute to distribute to users, then it wouldn't be the case. But on the case that it is, it would shift to a pay per usage model, which means that you buy credits and if you credits are consumed, you then need to buy more credits. It's kind of like the API model. Is that right? Yeah, according to the blog post, it says, from today through June 22nd, Fable 5 is included on Pro Max, Team, and seat-based enterprise plans and no extra

cost. On June 23rd, we'll remove Fabl 5 from those plans. Using it after that will require usage credit. 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. And yeah, it sounds like Fable 5 consumes a lot of compute. When you load it up inside of the app, it says Fable is the most capable model and draws down usage twice as fast as Opus. So you have to imagine that if you have to imagine that if It consumes a lot of GPUs. They are clearly using those GPUs for a lot of things. I think the idea is to give a preview and then extend that for as long as possible or just continue to

extend it perpetually based on compute. I think to your earlier point, we're very much at a fork in the road when it comes to these models being capable enough to really make a meaningful impact in the world. And we've spoken so much about alignment and AI safety. And it's kind of been this open-ended fuzzy thing where it hasn't really practically applied to anything that's happened before and we're finally at a moment in time in which the models are becoming capable enough to have that conversation about AI alignment, AI safety. You're starting to see why a lot of the teams are taking it so seriously because it is the singular question. It's like answering what you just said, who gets access to this model? How is it going to be restricted? Who gets to decide that?

And that's why the alignment and safety conversation is so important. And while you start to see a lot of the company cultures within these companies align around these different priorities that separate them from each other. So this is, it's, it's a new day. It's a new era today. We are moving into the next frontier of models. It was pushed forward a considerable amount in a way that I don't think we've experienced in quite a long time. And it's really exciting to see. I'm very excited to play with Fable, spend some time kind of generating outputs, figuring out what it's most capable in that could help us in the day to day. Like for me, if they never told me, it wasn't going to do bio or cyber, I'm not sure I'd ever come across it because

That's not really within the realm of uses that I have. So I'm excited to just kind of play with it and figure out best use cases for this. In terms of pricing, what I found really interesting is Cloudflable 5 is only twice the price of GPT 5.5. It's, I believe it's $10 per million input tokens, $50 per million out. And GPT 5.5 is $5 and $30 out. So pretty close and much more capable. So if the case that it does get removed from subscriptions, it is still available for my It is not as expensive as I think a lot of people thought.

This is significantly cheaper than what I believe Mythos preview was early on. Yeah, it's my new favorite model. I've been using it relentlessly for the last 10 hours. One thing that is a really strong capability that it has is long horizon tasks. Like this model is engineered from the ground up to be able to work like a dog for six to 12 hours at a time on whatever project that you have. And it has this loop function, which basically says, If you come across a problem, don't ask me, try and figure out yourself and do the thing, build the thing.

That's why we had people build world engines from scratch that we demoed earlier and these games from scratch all from a single prompt, the library of Babel. So if you have an idea or if you've been pondering on a project that you've been putting on for a while because you're like, I know I could probably vibe code this, but I don't want to spend like an hour doing this. Now you just need to write one detailed prompt and you should be able to do this. So my prompt for the listeners of the show as we wrap up this episode is get access to this model. Try it out. And I'm curious what your thoughts are on it. Do you think the restrictions affect you specifically?

Or do you think it's a really good general purpose model? And you're happy with how it's presented itself. You don't care about Metos 5 in effect. And also, what other projects are you going to be doing with this? Are there any kind of demos or use cases that we haven't covered that might be specific to you in your leisure or your particular work that you might want to apply this to. Let us know the feedback in the comments to this video or DM us on X.

Our profiles are linked below. We want to hear back from you. But I think that brings us to the end of this episode. We have now a new world-leading model. Open AI is going to have to answer to this, but it seems like Anthropic is running away with it. Josh, any final thoughts? That's it.

This is a new era today. I feel like we should celebrate. This is a new frontier that has been pushed forward very far in an industry we care very deeply about. So it's exciting to see. I'm stoked to use it. I'm curious to hear what the best types of prompts or use cases are that anyone who's listening has found. If you enjoy this episode, don't forget to share it with a friend who might also want to try Cloudfable 5 and experiment and get their feedback on how it's being used to improve their life, improve productivity,

whatever use cases it may be. But as always, thank you all so much for watching and we will see you guys in the next one.