Are Cheaper AI Models Better than Claude and ChatGPT?

We're discussing new AI model releases from xAI, Meta, OpenAI, and Anthropic, and the shift toward cheaper, more efficient models.Focusing on Grok 4.5 and Meta’s MuseSpark 1.1, there's also a broader move toward model routing and enterprise use.

------
🌌 LIMITLESS HQ ⬇️

EMAIL US:           [e-mail]
NEWSLETTER:    https://limitlessft.substack.com/
FOLLOW ON X:   https://x.com/LimitlessFT
SPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQ
APPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890
RSS FEED:           https://limitlessft.substack.com/

------
TIMESTAMPS

0:00 AI Price Competition
2:48 Grok 4.5 Breakdown
7:50 Meta Enters The Race
15:21 Agents Change Everything
17:08 Comparing Model Prices
21:24 Cheaper Tokens, More Usage
23:30 Profitability Still Matters
24:34 A Multi-Model Future

------
RESOURCES

Josh: https://x.com/JoshKale

Ejaaz: https://x.com/cryptopunk7213

------
Not financial or tax advice. See our investment disclosures here:
https://www.bankless.com/disclosures⁠

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.

Creators and Guests

Host

Ejaaz Ahamadeen

Host

Josh Kale

What is Limitless: An AI Podcast?

Exploring the frontiers of Technology and AI

Josh:
Just last week, in the span of about 48 hours, three of the most powerful AI

Josh:
labs on the planet all shipped brand new models.

Josh:
Elon shipped Grok 4.5, OpenAI took 5.6 Global, and Meta, for the very first

Josh:
time in history, put a price tag on its own frontier model.

Josh:
And here's where it gets interesting. For the last five years or so,

Josh:
the deal in AI was that every time a model shipped, it got smarter and cheaper at the same time.

Josh:
But that kind of died this year with Anthropics Fable 5 and these new frontier

Josh:
launches like GPT 5.6. So today, we're asking the question that probably everyone should.

Josh:
When frontier intelligence costs less than a cup of coffee, when they cost just

Josh:
a few pennies per millions of tokens, what does that look like in terms of your

Josh:
costs and how much you use these models?

Josh:
I mean, this is a totally different paradigm now. We have a very clear separation

Josh:
between Fable and 5.6 Sol and Grok 4.5 and Meta's new model, MuseSpark 1.1.

Josh:
And it seems like these models are kind of diverging in a way that's really

Josh:
interesting, particularly centered around price.

Ejaaz:
For the last couple of years, the ultimate validation of whether your AI model is good or not.

Ejaaz:
Is how intelligent it is, how smart it is. And no one really cared about cost.

Ejaaz:
And then Fable 5 kind of came on the scene. And I think it was like $20 or $30

Ejaaz:
input and like $80 output.

Ejaaz:
And companies that were spending tens to hundreds of millions of dollars started

Ejaaz:
to think, is this right? Does this make sense? Do I need the smartest model to do every single task?

Ejaaz:
And so we hear the likes of XAI, Elon's company. We look at the likes of Meta.

Ejaaz:
They released these models and we compare it to Fable 5, we're like,

Ejaaz:
they're not actually that smart.

Ejaaz:
But that's the whole purpose of the models that they're releasing.

Ejaaz:
They're not trying to be as smart as Fable. In fact, they're betting on the opposite.

Ejaaz:
They're betting that the cheapest model per unit intelligence is the model that

Ejaaz:
will ultimately fill the middle ground.

Ejaaz:
That'll be the ultimate model that is embedded in every single enterprise and

Ejaaz:
used by every single user.

Ejaaz:
Because the truth is you don't need the most intelligent model to do your task.

Ejaaz:
Maybe if it is 90 to 95% of the intelligence of the smartest model ever,

Ejaaz:
but it costs one-tenth of the price.

Ejaaz:
It's a no-brainer that you're using these models at scale. And like you said,

Ejaaz:
Josh, over the last week, there have been two particular companies,

Ejaaz:
Meta, who we've known and spoken

Ejaaz:
about a lot on this show, who have spent upwards of, I think, $35 billion.

Ejaaz:
To try and build the world's best model, came out with their new Meta Mew Spark 1.1.

Ejaaz:
And then you had SpaceX AI, who recently IPO'd, and there's a lot of pressure

Ejaaz:
writing on them, release their new Grok 4.5 models. Now, are they as good as Fable? No.

Ejaaz:
But are they cheap enough to use at scale when you're using or spinning up agents

Ejaaz:
or when you're trying to figure out that long, complex task that's going to

Ejaaz:
take tens of hours and you don't want to burn very expensive Fable tokens?

Ejaaz:
These models might be the ones to choose. And I think it's probably worth covering a bunch of them.

Josh:
Yeah, there's a new meta almost in town where it's like a new model doesn't

Josh:
necessarily mean higher intelligence.

Josh:
It could just mean higher efficiency. And I think that's where we're going to

Josh:
start with with the Grok 4.5 release, because this is kind of like an opus class

Josh:
claim, but at a third of the price, which is a pretty big deal.

Josh:
So this new Grok 4.5 model, it's built on XAIs or I guess SpaceX AIs,

Josh:
their version nine foundation model, which is about 1.5 trillion parameters.

Josh:
And for reference, the version we've been using all year, if you've used Grok

Josh:
at all this year, that is the version eight small, which is about 500 billion parameters.

Josh:
So we're looking at about a three times multiple in parameter count,

Josh:
which generally speaking is three times better, but probably a little bit more.

Josh:
It seems like this is a serious increase relative to what we've been using.

Josh:
And Elon has called this Opus class model. But instead of just being this like

Josh:
highly expensive, very slow model, it's much more quick and it's much more efficient

Josh:
when it comes to how many tokens you're able to generate for that same dollar.

Josh:
And this is very clearly the route that you could see SpaceX AI has been trying to go for a long time.

Josh:
They're very hardcore engineers. They love the engineering challenge.

Josh:
And what they're trying to do now is figure out how you can kind of sculpt these

Josh:
GPUs that they're training on to get as efficient as possible.

Josh:
We spoke a few weeks ago, Ijaz, about the etched guys, like the startup who

Josh:
is building their own training architecture chip stack. And they're basically

Josh:
building their own servers.

Josh:
And within that, they're pretty big to make one specific thing work,

Josh:
which is the transformer.

Josh:
And we talked about how gpus are not very efficient they don't actually use

Josh:
like sometimes up to 60 percent of the gpu isn't used what grok and the spacex ai team are doing

Josh:
is they are taking that code base they're really getting down to the bare metal

Josh:
to figure out how to squeeze the most juice out of it and that's what this model

Josh:
is that's what 4.5 is it has

Josh:
two dollars in six dollars out per million tokens generated

Josh:
And it seems like it's incredibly efficient when comparing it to other models like Opus.

Josh:
It appears as if it's up to four point times fewer tokens needed in order to reach the same task.

Josh:
So that's like an adjusted multiple of what it says, 17 times less than Opus.

Josh:
That's like a really big deal for a model as it relates to cost, at least.

Ejaaz:
The major unlock that you have with Grok 4.5 is it's a model meant for building.

Ejaaz:
So if you're a hobbyist out there, if you're a software engineer and you want

Ejaaz:
to tinker with some of these models and try and build something,

Ejaaz:
but you know it's costly and expensive. You don't want to use the Fable 5 API.

Ejaaz:
This is a really, really good model to use because it's so efficient.

Ejaaz:
So some of the stats you were referencing right there, Josh,

Ejaaz:
basically it uses four times fewer tokens than Claude Opus 4.8.

Ejaaz:
What that results in is a 17x less price or cost to do the same task.

Ejaaz:
And the way that they've been able to achieve this is arguably the funnest or coolest part.

Ejaaz:
They spin up a bunch of different agents to solve different problems of the

Ejaaz:
tasks that you've asked it to at the same time in parallel.

Ejaaz:
And this is a growing trend you'll see with the other models that we're going

Ejaaz:
to talk about on this show today, where agentic coding or agentic reasoning

Ejaaz:
or using agents to solve your problem has been a major unlock for these cheaper

Ejaaz:
models to be cheap in the first place.

Ejaaz:
The other thing I want to mention about Grok 4.5 in particular,

Ejaaz:
because it's very unique to them and SpaceX AI is they just acquired a company called Cursor for...

Ejaaz:
You know, a low price of $60 billion. And the major unlock that cursor gave

Ejaaz:
them is they have all this data around how users use coding models.

Ejaaz:
Now, it's very specific. It's not what they're coding. It is how they use the models.

Ejaaz:
And this data was very important in training Grok 4.5 to become smarter at routing people's requests.

Ejaaz:
So let's say you ask it to build an app that can change your wardrobe into something

Ejaaz:
way better. And you send it like pictures of your camera or whatever that might

Ejaaz:
be. We talked about this as an example on yesterday's episode.

Ejaaz:
It'll be able to know which model to use at what time for how long and calculate

Ejaaz:
the costs preemptively to make sure that it's not burning your wallet.

Ejaaz:
It's a really, really smart model. And Elon himself has said,

Ejaaz:
listen, this isn't gonna be the smartest model. We're working on better models,

Ejaaz:
but it is a really good daily driver. It's a really good workhorse.

Ejaaz:
And if you're anyone at a company that doesn't want to burn, you know,

Ejaaz:
$30 in and $80 out on Fable 5 or whatever the cost is, something that's like

Ejaaz:
10x more, you should use this model for 80 to 90% of the work that you're trying

Ejaaz:
to do, and then use Fable to kind of orchestrate the plan or the design or whatever that might be.

Ejaaz:
The final point I'm going to make in SpaceX AI's favor, because you might be

Ejaaz:
thinking, okay, fine, whatever, but like, when's the next model going to come?

Ejaaz:
They took like, whatever, nine months to release this upgrade.

Ejaaz:
Elon is cooking up three more models that are in order of magnitude larger than

Ejaaz:
the model that we're talking about today.

Ejaaz:
So right now, in under a month, we're expecting to see Grok 5,

Ejaaz:
and Grok 5.5 is also being cooked at the same time as well.

Ejaaz:
And these are like 5 to 10 trillion parameter models. So it's feasible to say

Ejaaz:
that we'll have a Mythos class model from SpaceX AI in a couple of months' time.

Josh:
Which is awesome. Yeah, this is the part that's most exciting to me.

Josh:
Is like it's very obvious that spacex ai is the best at the engineering part

Josh:
of generating tokens they're

Josh:
have the whole vertical integration now they have the ability to build out these

Josh:
data centers they have the data centers running and now they have the harness

Josh:
through cursor and they also have the data set that they've used through cursor in order to kind of

Josh:
achieve this fully integrated stack and what's funny now is if they are continuing

Josh:
to release models that are bigger and bigger

Josh:
if they ever do run up against a compute wall they remember they have that deal

Josh:
with anthropic and with google i believe and

Josh:
they're gonna have to figure out who's who's gonna get the short end of that

Josh:
stick because it doesn't seem like they're gonna have gpus for everyone but

Josh:
like you mentioned the the exciting thing here is that like this is version

Josh:
one of it feels like spacex ai 2.0 this is their

Josh:
second try at getting to the frontier the first time they may have gotten there

Josh:
for a couple days but it wasn't very long lived now we're getting new models

Josh:
every single month and the goal

Josh:
around august is a two trillion parameter model and then the goal for grok 5

Josh:
which is coming hopefully not too long after is 6 to 10 trillion parameter models

Josh:
like this is this is going to put them right up at the frontier and if they're

Josh:
able to serve these tokens

Josh:
at a fraction of the cost that say gpt6 is going to be or mythos 6 or whatever

Josh:
comes next that's going to be a pretty serious

Josh:
competitor in the ai space because they're going to be right at that frontier

Josh:
with a very low cost model that i think a lot of people

Josh:
Are going to find a lot of use for now that is the grok update there is a second

Josh:
update that is just as noteworthy and probably even more noteworthy actually

Josh:
because this is the first time that meta is charging for tokens

Josh:
meta famously they have been the open source kings they have always wanted to

Josh:
publish the models open source to move the needle forward to kind of make this

Josh:
open source developer community thrive

Josh:
unfortunately if you are a participant of that your time has ended because now

Josh:
meta is closed source and they are releasing

Josh:
these closed source models they are charging via the api it's not a lot of money

Josh:
but it was big enough news for mark zuckerberg to come back on x after a what

Josh:
was a three or four year hiatus and actually announce

Josh:
muse spark 1.1 which he describes as a strong agentic and coding model at a

Josh:
very low price it's available through our new meta model api and in the meta

Josh:
ai so ejes the question i have for you

Josh:
because you know i'm i haven't been the most excited about meta recently

Josh:
their offerings have left a lot to be desired as it relates to AI.

Josh:
Is this a serious model? Like, is this worth actually paying for relative to

Josh:
all the other models that exist today?

Ejaaz:
Short answer is yes. And this comes from a professional meta hater.

Josh:
Which is shocking. Like this is this is a novel breakthrough.

Josh:
This is an exciting announcement for meta.

Ejaaz:
Well, I'm just happy to see something competitive enough for the,

Ejaaz:
a lot of people to use in the AI market. And I'll tell you why they would use this model.

Ejaaz:
Number one, the reason right at the top is it is the cheapest model per unit

Ejaaz:
cost of intelligence. So what do I mean by that?

Ejaaz:
You can have cheap models, but they're pretty crappy and you won't end up actually

Ejaaz:
using it for serious work.

Ejaaz:
This is a model that you'll end up using for serious work and it costs 25% of

Ejaaz:
the price of the Frontier model. Now, can it do everything a Frontier model

Ejaaz:
can do, like what Fable 5 does? No. And Mark Zuckerberg openly admits that.

Ejaaz:
But it is an absolute workhorse. You can throw it at a task and it can work

Ejaaz:
for hours, or you could throw it at multiple problems at once and it can figure it out.

Ejaaz:
But there's a few other advantages that this model in particular has.

Ejaaz:
So aside from the price, which is, by the way, $1.25 in per million tokens and $4.25 out.

Ejaaz:
Now, if you want a comparison as to how crazy cheap that is,

Ejaaz:
that is cheaper than GLM's model, GLM 5.2, which is the leading open source

Ejaaz:
Chinese model right now.

Ejaaz:
And they're known for being the cheapest model. So the fact that Mark Zuckerberg,

Ejaaz:
there's some blissful irony there, actually, with the open source thing and

Ejaaz:
the fact that, you know, he was competing with China and they beat him.

Ejaaz:
To come back and offer the cheapest model is pretty amazing. But the second thing is.

Ejaaz:
And like Grok 4.5, it uses agents very intelligently. So it has this thing called

Ejaaz:
a master agent in this model.

Ejaaz:
And the master agent reads your prompt and it thinks very diligently about a

Ejaaz:
plan to answer and execute your prompt.

Ejaaz:
Now that sounds very vague, right? You're like, oh, aren't the other models

Ejaaz:
doing it? No, when you look at like Fable 5, when you look at GPT 5.6,

Ejaaz:
it processes its entire model weights, which is gargantuan by the way.

Ejaaz:
And that ends up being very costly. Meta found a sneaky little way to circumvent

Ejaaz:
this, which is have a master agent, have it plan, and then have it delegate

Ejaaz:
to a bunch of different agents.

Ejaaz:
The third thing that is very impressive about this model is that it's an omni model.

Ejaaz:
So it can take in video, it can take in images, it can take in text all at once

Ejaaz:
and understand how to use that intelligently.

Ejaaz:
So if you give it a video, you can extract that video, it knows that you want

Ejaaz:
to list it or use it for a post that you have on Facebook Marketplace,

Ejaaz:
and it'll be able to do that in one shot.

Ejaaz:
And then the fourth and final thing that it's very good at is computer use.

Ejaaz:
So this thing can take over your computer, take over your account or whatever

Ejaaz:
that might be, and just know what you want it to do. Now, version one is very

Ejaaz:
fine tuned to Meta's products, Instagram, WhatsApp, Facebook itself.

Ejaaz:
But the idea is you can pretty much use this for any computer use work going

Ejaaz:
forward. And I'm looking forward to version 1.2 and 1.3.

Ejaaz:
They pulled this off, by the way, within a month of releasing Mew Spark 1.

Ejaaz:
And they also released Mew's image and video in the same week,

Ejaaz:
which tells me that the cycle of release similar to SpaceX AI is getting much quicker.

Ejaaz:
And if you ask me why that is, it's because both Elon and Zuck have one thing

Ejaaz:
that the two frontier labs, OpenA and Anthropic, don't have.

Ejaaz:
A crap ton of compute. They have so much compute. They are the most aggressive

Ejaaz:
hyperscalers and arguably the most successful hyperscalers. They have amassed the most GPUs.

Ejaaz:
And if you still believe that compute scaling laws matter, Zuck and Elon are

Ejaaz:
not out of the race. In fact, this proves that they're back into it.

Josh:
One of the things that I found noteworthy of this is that this API presumably

Josh:
runs on Meta's own silicon.

Josh:
It's the chips that they have been making and producing and likely running in

Josh:
fact there's a report that they are spending 250

Josh:
billion dollars including chips over the next year in order to build out something

Josh:
like 14 gigawatts of data center capability

Josh:
and that is mostly going to be running their new mtia 400 chips which are basically

Josh:
the in-house meta silicon which is 400 faster than the previous generation and

Josh:
uses 51 percent more hbm so for the hbm folks

Josh:
When we talk about all our investing videos um these chips are going to be using

Josh:
a lot more of it and meta is now charging a quarter of the rival's prices which is a really

Josh:
kind of interesting and noteworthy thing and that combined with the switch from

Josh:
open source to closed source it kind of implies that like mark zuckerberg very

Josh:
clearly thinks the model is now the product instead of the moat around

Josh:
the model so i think traditional meta would have been no no we're going to release

Josh:
the model we're going to build a moat around it and that is going to be product we monetize

Josh:
this new shift implies like no actually the model is the product and we're going

Josh:
to integrate it into all of our services but in order to do that we're going

Josh:
to do this huge tremendous data center build out and they're going to

Josh:
build their own proprietary chips and it seems like both those things are actually

Josh:
going well and i have to ask it's like okay

Josh:
what happens if they actually do it like what happens if there is 14 gigawatts

Josh:
of compute running next year like that's that seems pretty considerable it's

Josh:
particularly running on their own silicon

Josh:
which we know has tremendous competitive advantages when it comes to training your own models.

Ejaaz:
Well, here's the bet that like we're making, we've made on this show multiple

Ejaaz:
times on previous episodes, is that the future of AI isn't you tapping a bunch

Ejaaz:
of buttons and approvals every single second.

Ejaaz:
It is you write a prompt, you send the prompt, and then the AI just kind of

Ejaaz:
knows what to do. It autonomously works.

Ejaaz:
If you believe in that world, then you believe in a world of AI agents.

Ejaaz:
And if you believe in a world of AI agents, these agents are going to be making

Ejaaz:
tens to hundreds of thousands of tool calls per month.

Ejaaz:
You don't want to be there clicking approve the entire time.

Ejaaz:
And those tool calls are pretty expensive.

Ejaaz:
So you want to go the cheapest route when you're using an agent to get the work

Ejaaz:
done. That's basically the entire thesis for why cheaper models are more effective.

Ejaaz:
And therefore, the labs who have the most compute and can use it most effectively,

Ejaaz:
you mentioned, you know, their custom chips.

Ejaaz:
I believe Elon and SpaceX are also working on their own chips.

Ejaaz:
Open Air is doing their own with Jalapeno. This is a growing trend.

Ejaaz:
It makes sense that the people that have the most compute and the best chip

Ejaaz:
architecture will end up winning. And that's basically the bet that Meta and Elon are going after.

Ejaaz:
I have to say, if we ground ourselves for a second and look at the counter-thesis...

Ejaaz:
I do think Zuck is heavily subsidizing this model.

Ejaaz:
I don't think there's any chance in hell that it actually costs $125 and $425

Ejaaz:
output. I think he's subsidizing this massively.

Ejaaz:
And I think he needs to prove a point to his investors or shareholders that

Ejaaz:
it is worth the AI CapEx spend that he's probably going to announce at the end of Q2.

Ejaaz:
So that's my bet. I think it's strategic. I think it's the right move,

Ejaaz:
but I don't think he's unlocked some kind of major architecture redesign just yet.

Josh:
Regardless, these are two really solid models between Grok 4.5 and now MuseSpark 1.1.

Josh:
And to kind of place them in a spot relative to others, we can go down the price

Josh:
list of other models and kind of compare what they're like. So at the top of

Josh:
this list is Claude Fable 5, and that's $10 in, $50 out per million tokens.

Josh:
GPT 5.6 comes next at just a little bit, close to half at $5 in, $30 out.

Josh:
Then Opus is $5 in, $25 out.

Josh:
And then it kind of goes down the line until we get to Grok 4.5,

Josh:
which is two dollars in six dollars out then we have muse spark 125 and 425

Josh:
and at the very bottom believe it or not is open ai with a dollar

Josh:
in and six dollars out for gpt 5.6 luna so the middle section and the upper

Josh:
section is kind of where the war is it's like we have those two models at the top

Josh:
we have claude and we have gpt 5.6 soul

Josh:
Those are very much competing on the intelligence curve. But then below that,

Josh:
where we have this like kind of cluster of, you can think Gemini and the smaller

Josh:
GPT models and Grok and Meta 1.1, that's where we are seeing this like price competition.

Josh:
And it's funny to see this kind of divergence in strategies.

Josh:
And I think that's a good way of looking at the frontier when you evaluate these

Josh:
new models is, okay, is this a frontier intelligence model or is this a frontier price model?

Josh:
And those two things now are very different because they're going to be used

Josh:
for a very different set of use cases. And I think one of the more interesting

Josh:
applications that we're going to be following on the show is how people route

Josh:
through these models to do different tasks.

Josh:
Like you said, some agentic tasks don't require necessarily the highest frontier intelligence.

Josh:
Maybe you can get away with using a MuseSpark 1.1 for a lot of that,

Josh:
and perhaps using a Fable 5 for orchestration of those agents, things like that.

Josh:
So we're going to see what I suspect is a new meta start to come into play of

Josh:
this orchestration at a high level and using different models for more particular

Josh:
things, more specific things.

Ejaaz:
Josh, have you heard of something called the Silicon Token Expenditure Index?

Ejaaz:
It's basically this index which tracks the spending of companies or enterprises.

Ejaaz:
I think the Bloomberg ticker is SDLLMTK.

Ejaaz:
But the point of this index is it captures how much money is being spent by

Ejaaz:
the top Fortune 500 on AI specifically.

Ejaaz:
So if you're listening to us talking about cheap models and And,

Ejaaz:
you know, talking about this thesis of cheaper models will be used more and you don't believe it.

Ejaaz:
Well, you have to look at the customers who are actually ingesting this AI and

Ejaaz:
the movements and actions that they're taking.

Ejaaz:
And if I show you the index, you'll notice a particular trend over the last

Ejaaz:
couple of weeks and months, which is it's down 20%, which basically means people

Ejaaz:
are spending much less on tokens or per token.

Ejaaz:
And they're also spending much less overall. What that indicates is they are

Ejaaz:
looking for cheaper alternatives.

Ejaaz:
And we're seeing this from the headlines that we've seen from Uber,

Ejaaz:
from Meta themselves, from Microsoft, who are now adopting Chinese open source

Ejaaz:
models into their product, into their Copilot product.

Ejaaz:
We are seeing this shift of enterprises realizing that it's not about using

Ejaaz:
the smartest model for every single task.

Ejaaz:
It's about finding that middle ground. It's about finding the right model or

Ejaaz:
maybe the right types of models to use for the middle ground of tasks.

Ejaaz:
And that brings me to another point which is.

Ejaaz:
I don't think it's necessarily just going to be one cheap model that dominates everything.

Ejaaz:
I think they're going to realize after they've shifted to a cheaper model that

Ejaaz:
they could use certain models for specific tasks.

Ejaaz:
And that takes us down the path of routing, which is what Cursor has infamously

Ejaaz:
figured out and what SpaceX figured out and acquired them for $60 billion.

Ejaaz:
So there's these really interesting conversations around this trend,

Ejaaz:
but it's being validated by this index that companies are going to be looking for cheaper models.

Ejaaz:
And the irony is, what is the paradox, Josh, that we spoke about early on in

Ejaaz:
the AI thesis, where the cheaper something gets, the more you end up spending

Ejaaz:
on it? What is the name of that paradox?

Ejaaz:
Jevon's Paradox. Jevon's Paradox. That's it. So the cheaper something gets,

Ejaaz:
I expect to see way more tokens being spent because the output will actually

Ejaaz:
be worth it instead of spending a very expensive prompt and getting maybe a

Ejaaz:
mediocre answer from it.

Josh:
Yeah, and this is in line with expectations. Goldman, they published this prediction

Josh:
that token consumption would multiply 24 times between 2026 and 2030.

Josh:
That's like a tremendous amount. That's 120 quadrillion tokens per month, which is unbelievable.

Josh:
So even if the price of these tokens does go down some extent,

Josh:
a multiplier of that much is still increased spend, which I think is important to note.

Josh:
It's like, I don't think we're going to see a Fable class model be 124th the price very shortly.

Josh:
But they're expecting that much increased demand and every single kind of

Josh:
Every single prevailing wind is pushing towards longer-term agentic coding sessions,

Josh:
where I feel like rarely do people actually now, or people who are using it

Josh:
for productive uses at least, are using it just as a chatbot.

Josh:
They're using it as an agent to do longer and longer and longer-term tasks.

Josh:
And what we notice with these cheaper models in particular is that they actually

Josh:
do oftentimes consume more tokens to get the same output at a higher quality.

Josh:
I know we were mentioning this with 5.6 the other day, where some of their lower

Josh:
models, they actually, they cost less, but they do use a considerable amount

Josh:
more tokens to get to the answer.

Josh:
So we're in this weird crossroads where we have these, like,

Josh:
forcing functions that are pushing people to want to generate more tokens.

Josh:
Companies want these tokens to be highly intelligent because they don't want

Josh:
subpar work if they're paying for it.

Josh:
But then they have this like enterprise belt tightening that's kind of happening

Josh:
where like amazon killed it's like i think they had their internal token leaderboard

Josh:
and they totally removed it

Josh:
um coinbase and walmart are setting up usage caps a lot of companies are kind

Josh:
of like crunching down on usage of ai internally because they can't really figure out how to justify

Josh:
the value that's been given to the company so there's a lot of these weird it

Josh:
very much feels like we're at a crossroads now and there's this new paradigm

Josh:
shift that is happening as it relates to cost in particular and it's going to

Josh:
be really interesting to see how this plays out

Ejaaz:
There is one metric that i will be tracking to see whether these cheap models

Ejaaz:
are actually good enough and it's simple,

Ejaaz:
are these companies making money? When do they turn profitable?

Ejaaz:
With Meta, I'm almost certain that they're subsidizing their cost of their model.

Ejaaz:
With SpaceX AI, I don't know, maybe they're doing something similar.

Ejaaz:
But until these guys post a positive revenue earnings on their quarterly earnings

Ejaaz:
or whatever that might be, I'm not going to be convinced that cheap models are the way forward.

Ejaaz:
I think it will directionally play out, but that's going to be a big metric

Ejaaz:
to track. Now, if you compare that to the likes of Anthropic,

Ejaaz:
they're not a public company yet. They're rumored to IPO maybe later this year.

Ejaaz:
But there are rumors that they have turned profitable in Q2.

Ejaaz:
And if true or if confirmed, they'll be the first AI lab to do so.

Ejaaz:
And the major way that they've been able to do that is not only creating amazing

Ejaaz:
models that are used by every single enterprise in the world,

Ejaaz:
but because it's expensive, because they pay for the cost that you're getting.

Ejaaz:
And at the end of the day, if you're making a lot of money and people continue

Ejaaz:
using it, maybe that is the right way to go through. So that that would be the

Ejaaz:
only the counterpoint to this.

Ejaaz:
The final thing I'll say on this topic, I think, is.

Ejaaz:
What I was alluding to earlier, which is I'm now convinced that,

Ejaaz:
number one, it's not just going to be OpenAI and Anthropic ruling the entire world.

Ejaaz:
I think it's going to be multiple labs. And I think that's ultimately a very

Ejaaz:
good future, right? Maybe five, six, maybe 10 labs, right?

Ejaaz:
And I think they're going to be so many different models to use for so many

Ejaaz:
different purposes. I mean, look at Metamuse Spark 1.1.

Ejaaz:
It topped the health benchmark. Did you know that?

Ejaaz:
Like a random social media company's model topped the health benchmark.

Ejaaz:
Grok 4.5, really good at computer use and spinning up agents.

Ejaaz:
Luna from OpenAIR, also really good at computer use. So all these different

Ejaaz:
models would be good for various different tasks.

Ejaaz:
And if you don't want to spend time thinking about which subscription to get

Ejaaz:
or what model to use or when, I think routing companies.

Ejaaz:
So what cursor basically built, which is you can type in a prompt and they decide

Ejaaz:
which parts of the prompt gets used by which kind of models will be the killer platform or product.

Ejaaz:
I don't know of any company that is building this that hasn't been acquired, aka Cursor.

Ejaaz:
Maybe OpenRouter. We've had the founder, Alex Atala, on our show before.

Ejaaz:
Maybe they pivot into some kind of a routing product. But I'm convinced now

Ejaaz:
that it's going to be a layer that sits on top of these models, potentially.

Josh:
Yeah. I think the way I'm looking at it is like, okay, now there's two markets.

Josh:
This used to be a singular commodity, a singular race to the top.

Josh:
Now there's these two things. And like cost per intelligence is the right lens

Josh:
for looking at maybe 80% of the work that's become like kind of routine,

Josh:
but it is the opposite and exactly the wrong lens for like the 20% where

Josh:
capability and reliability and verification matter.

Josh:
It's like if you are doing really complicated work where mistakes cost you a

Josh:
tremendous amount of money and time, if you're doing frontier math or problem

Josh:
solving, or you're building complicated things as a business,

Josh:
there's almost no limit to the amount that you'll spend in order to accomplish your goals.

Josh:
Because I'm sure many of these companies have a tremendous amount of ideas that

Josh:
they want to implement to the market.

Josh:
And the constraint is their workforce and knowledge base that could actually deploy that.

Josh:
There is no limit to the amount that they will spend on these Fable models,

Josh:
on these GPT 5.6 and GPT 6.0 models.

Josh:
But for the rest of the world, maybe that's not the case. They are price sensitive.

Josh:
They don't need the frontier intelligence for everything.

Josh:
And that's where a lot of these models are going to really play a serious role.

Josh:
And i think that is cool that like we no longer have intelligence as the singular

Josh:
commodity it is now split between intelligence and price those two things both

Josh:
matter and i think that's kind of where we are in the market right now it's

Josh:
like we are at this crossroads the commodity has diverged into two

Josh:
and we have some pretty serious players in the pricing game who have a very

Josh:
serious trajectory of actually continuing to be serious threats spacex ai is

Josh:
building a huge amount of data centers they have a very clear trajectory to a

Josh:
20 trillion parameter model meta is going to put 15 gigawatts on the ground power it on

Josh:
and they're going to do that vertically integrated with their own silicon and

Josh:
like that's a pretty serious threat to the pricing war because you have to imagine

Josh:
they're going to get some

Josh:
amazing efficiency from that so yeah i think that's the update for

Josh:
The state of low-cost models. It's been a very interesting week for them.

Ejaaz:
As we're recording this episode, I'm realizing that we need to make another

Ejaaz:
episode on these trends that we've spoken about during this talk.

Ejaaz:
Primary ones being, okay, if the most expensive model lab or the most intelligent

Ejaaz:
model lab isn't necessarily going to win over the next couple of years and it's

Ejaaz:
going to be this middle ground, this cheap middle ground, which companies are

Ejaaz:
there? Where are the bottlenecks?

Ejaaz:
How are people using these things?

Ejaaz:
What will agents look like when they're working on long horizon tasks?

Ejaaz:
And most importantly, which companies, which customers are going to be using

Ejaaz:
this thing? I think, I mean, you let me know, as you as the audience that are

Ejaaz:
listening to this right now, would that be an interesting episode to sort of unpack and dig into?

Ejaaz:
All of that said, I think that cheaper models are here to stay.

Ejaaz:
And I think they're here to stay in a very big way. And I think they're going

Ejaaz:
to come out, thankfully, from some frontier Western labs.

Ejaaz:
We're shifted away from building the most expensive and intelligent model that

Ejaaz:
is being kind of blocked to anyone and everyone to a $1.25 in,

Ejaaz:
$4.25 out model from Meta, which surprisingly works and is intuitive.

Ejaaz:
I'm also excited to see what people do with these cheap models.

Ejaaz:
I don't think they're necessarily going to build the same things that people

Ejaaz:
using Fable 5 are going to do simply because they're not prohibited by cost anymore.

Ejaaz:
And I think that kind of unlocks a new line of creativity when you're building

Ejaaz:
these apps. You're like, oh, you know what? Maybe I will try this crazy thing

Ejaaz:
because it's only going to cost me 10 bucks.

Ejaaz:
Heck, I'll just do it. So if you're listening to this.

Ejaaz:
And you're inspired by some of the things that were spoken about,

Ejaaz:
go and try some of these models. They're super cheap to use.

Ejaaz:
And tell us what you end up building with it. Tell us if there's anything creative

Ejaaz:
that you haven't seen being built by Fable 5.

Ejaaz:
We would love to see it, demo it, heck, maybe even show it on an episode in the future.

Ejaaz:
But if you enjoyed this, please, please, please share this with your friends.

Ejaaz:
If you're not subscribed, please subscribe.

Ejaaz:
If you haven't left us a comment, we always want to hear from you.

Ejaaz:
We love hearing from you guys.

Ejaaz:
And if you're listening to us on Spotify or Apple Music, please,

Ejaaz:
please give us a rating. it helps us out pretty massively josh any any final thoughts.

Josh:
No that's it thank you guys so much for watching um on sponsor front we're chatting

Josh:
with some people this has been going well so for the people who have reached

Josh:
out thank you we are working our way through those we are

Josh:
slowly working towards becoming a self-sufficient entity which has been amazing

Josh:
so thank you so much for the support on that again if you know anyone please

Josh:
refer them send them our way

Josh:
either on x email and description whatever it may be

Josh:
but that is another episode next coming up is the roundup which is i think our

Josh:
favorite episode of the week we just throw everything that we haven't been able

Josh:
to talk about into one and do a full recap on everything that has happened this week which is a lot

Ejaaz:
We got to talk about the data center bands dude like new york is my city why are we banning the day so.

Josh:
We can have a lot of conversation about that one so stay tuned stay tuned for

Josh:
that grieving session um but anyways that is the episode today thank you all

Josh:
so much for watching as always and we will see you in the next one