Anthropic just dropped Claude Opus 5 and

it is by far probably the best Frontier

large language model currently available

to us underassmen. In this video, I'm

going to show you everything that you

need to know about how to take the most

advantage and use Opus 5 for the best of

its capabilities. This is a website that

essentially Opus created itself. It's a

full 3D application that allows me to

kind of jump in, see the cgraphy of

nowhere. Really what this is is it's a

3D world and uh I was able to simulate

this and create this entirely in Opus 5

with virtually zero work. So what we've

done here is we've essentially created a

bunch of art. We've then put these art

uh you know these these things up on the

wall and I'm just walking through it.

Every time I mouse over you can hear

this kind of like a little ding. That's

uh you know me looking at this work and

and essentially cataloging it like a

Pokédex. So this is not easy stuff to

do, right? I mean, it's not a long time

ago that this thing would have been

considered a full game and sort of 3D

experience in its own right. I bet you

if I asked Opus 5, it could turn this

into an actual virtual reality

experience in a few seconds, complete

with, you know, enemies and hens and

lasers and whatever the hell else I

want. Well, I could show you benchmarks,

and I will in a second. Suffice to say,

Opus 5 kicks ass on virtually all

benches. And I'm going to break down

exactly what that looks like in a

second. But I think the more important

thing here at this point is how does the

model feel? What sorts of outputs can it

generate that uh you know I can rate

based off taste, not necessarily little

static percentages on a screen that

let's be real don't really mean anything

to us anyway. So what follows is a

massive list of all the impressive

things I got Opus 5 to do. The things

that I consider visually stimulating and

also pretty interesting. Over here is

pretty neat. This is like a 3D

essentially Kerbal Space Program style

launch game. It allows you to build a

stable orbit in by changing the

trajectory of this. So if I if I launch

something, let's say from the launch

site in that direction, what I can do is

I can actually have it enter essentially

the gravity of the planet, which is

pretty badass. Obviously, this is kind

of on the simpler design side, I would

say. But it's still kind of neat that I

could just ask AI to do this, and I sort

of have my own simulated solar system.

It really does feel like simulations are

where all this stuff is going. There's

Newton's cannon over here which actually

shows a bunch of these launched. And

then we also have Hoffman with a big

target orbit over here. We're

essentially trying to make this this

thing. So, it's both an explainer and

it's also kind of a miniame. This is a

simulation of fabrics and how they move

in the wind. So, it's sort of a cloth

simulation. And you can see I I'm even

capable of ripping this simulation

apart, which uh is very neat. You know,

I can't say that this uses fabric

dynamics, you know, about as accurately

as possible. But I mean, you know, when

I'm looking at this and I'm thinking,

damn, this is pretty accurate. Uh, this

feels like it would respond the same way

that, you know, a curtain or something

would respond if I were to pull it. How

about this fractal generator? I mean,

this thing is insane. You can zoom in

and you can go, I mean, however deep you

want with this shader. It's honestly

quite ridiculous. You can also see that

it's sort of pulsing and growing. And

you know, we can increase the height and

the size, the bloom, how much it sort of

shines, dispersion, and so on and so

forth. Speaking of simulation, I don't

know if you guys have ever played this

game, but this is essentially a

collection of cells that are growing in

a tube. Every time I press this button

down, what I'm doing is I'm dropping

some uh sand down. So, it allows Opus 5

to design this environment where these

cellular automatons are sort of

climbing. I can also do, I don't know,

cause an oil spill here if I really

wanted to or maybe generate some steam

underneath that, um, I don't know,

destroys a lot of that water and stuff

like that. Um, so this is pretty badass.

I mean, it's fun that I can do that. I

can also cause fire. I can regenerate

starter scenes. I can stop time. I can

do a lot. This over here is like a

living predator prey ecosystem where I

can spawn new bunnies. These bunnies can

then start consuming, you know, grass in

the environment. But then there's also

what looks to be some foxes that are

hunting. And so, you know, I just caused

a bunch of bunnies to appear. Let's get

a bunch of foxes to appear. Thin the

herd a bit, shall we? And you can see

what's happening is now there's a

massive prepoundonderance of grass here

cuz the fox have just eaten all of them.

However, the fox population is now

crashing cuz there just aren't a lot

more bunnies available to eat. How about

this quadcopter flight simulator? I

mean, here we are with my little

quadcopter drone. I don't know if you

guys could hear, but the drone itself is

literally making sound. And I'm just

proceeding through my totally, you know,

procedurally generated environment. It

looks like if I wanted to land, I' i'd

hold X. So, I'm just going to try that

and then land my helipad right there.

Nice. And I just did. So, this is 100%

like a game. And yeah, just coded that

in a few seconds. This one's a

double-sided pendulum. I'm sure you guys

probably familiar with what one of these

do. Uh, interestingly enough, it does

kind of look like it screwed up the

design in the top right hand corner,

although that is the first screw-up that

I've been able to see so far. You can

add as many pendulums as you want and

like redo this. Nothing super special

here. Let's move on to something cooler

here. It designed a pixel editor, so I'm

actually capable of drawing just like

something in Microsoft Paint. Uh, we're

very close to everybody being able to

design their own sort of custom

applications. But yeah, you can see how

it has all the functionality. Has a

little eraser functionality. I can, you

know, undo whatever I want. I can, I

don't know, do some sort of onion skin

or even multiple layers. So I can add an

additional frame and then what I can do

is I can actually like make a movie

where I have this be my first frame,

this be my second frame all the way up.

So that's kind of neat. Hey, you can you

can sort of forge something. You can

even make the canvas bigger or smaller,

which is kind of neat. And you see when

I made it smaller, it does not look very

happy with here. We're building sort of

a sneaker configurator. So I'm not going

to say this is the clearest and cleanest

example of a sneaker I've ever seen, but

I don't know. Looks like I can click on

various things and change a bunch of

this. So, let me just shuffle sort of

change the design up. Maybe I want a new

color here. You know, that looks nice.

That looks like maybe it's like a Nike

style thing. I can change the material

and that'll slowly change what's going

on. I now have liquid chrome down. Looks

uh like I can also save a render to my

computer, which is kind of neat. Or just

stop it entirely. I can then move this

thing around, click on different

sections if I wanted to change things

custom. Not going to lie, not getting a

shoe vibe here, but I think the reason

why is cuz Opus 5, it didn't just pull a

shoe from the internet. What this did is

it actually created these elements

itself, which is a big step up because

that sort of compositing is usually not

what models do. We have what looks to be

here a explainer graph showing the

populations of different centers over

the course of the last 126 years. So in

the top right hand corner, you can see

the populations of New York change,

London change, Tokyo change, and so on

and so forth. while it explains it to

us. Um, so now we're moving over to Asia

and we're seeing that postwar Tokyo is

now at 16 million. And uh, you know, I

can also drag this and move this however

I want with populations just getting

increasingly interesting, I should say,

as we go. Can also zoom in, zoom out, do

all this fun stuff. Okay, this one's

interesting. This is a wrecking ball

simulator. And so I can actually drag

this, move this around, and then push

this towards this [laughter]

brick wall. And I'm not doing a very

good job, mind you. Definitely not

coming in like a wrecking ball. But

yeah, you can see that uh that this is

falling, which is pretty wild. You can

also change the mass of the ball, make

it heavier or lighter, which is kind of

fun. Um anyway, let's pretend I have a

bunch of glass here.

There you go. And there's even a little

like slider progress bar that's asking,

hey, you know, how how close am I to

destroying the entire thing? Apparently

96%. So this thing's pretty close. Yeah.

I mean, this can virtually do anything

that you would ever want it to do. And

if it's not clear, uh, the reason why

I'm so excited about this is not because

it can do all these things. Obviously,

there are a lot of models out there now

that can do things like this, but I

think the way that it does the things is

a little bit better than those models, a

few percentage points better anyway. And

I think more importantly, it's also way

cheaper. And I'm going to show you guys

this in a second, but this is far

cheaper than any of the other major

large language models, especially the

ones at the frontier, um, for these

sorts of tasks. So, what's my take on

this whole thing? Uh, Opus 5 is

basically slightly more intelligent

Fable at maybe half of Fable's costs.

And so, we see the cost of this website

here, Airlines, which is sort of like a

a Mac or Apple style website apparently

trying to sell you some sort of 3D

panels um was 69 whereas Fable 5's was

94 cents. And you can see that I

actually think Opus 5 did a better job.

I I don't think the websites are really

even comparable. If I open both of them

up in new tabs, this is the one that

Opus 5 just made. Okay, where there's

this cool kind of slide thing, keeping

in mind that this is not, you know, a

resource downloaded from the internet.

It just made these resources with SVG.

And then this is Fables over here,

which, you know, I think is just way

simpler and probably a lot dumber in

reality. So, on to benchmarks. You could

see Opus 5 here does pretty darn well on

agentic terminal coding. So much so that

it's actually considered better than

Fable 5 at the same task. and far better

than Opus 4.8 and even GPT 5.6 Soul

which scores a little bit above Fable 5

at least as of the time of this

recording on knowledge work on GDP val-

AAV2 scored 1861 so this puts it

squarely above like the average human at

knowledge work um at novel problem

solving using arc AGI 3 it scored 30.2%

Fable has not actually used this

benchmark yet so there's no stat here

opus 4.8 and it scored to 1.5%. So, as a

progression from, you know, Opus 4.8 to

Opus 5, at least in the Opus series,

this is a massive step up. Um, on

Agentic Search, you see that it scored

90.8% on browse comp, a little bit

better than Fable 5, significantly

better than Opus 4.8, but essentially

the same as GBD 5.6 Soul.

multi-disiplinary reasoning it scored

56.3% with no tools but when tool

augmented it actually outperforms fable

at 64.7 to fable 63.9. So that'll be

very useful um especially for longstep

sort of human knowledge work tasks.

Computer use was 70.6. So we're seeing

models get better at just using a

computer the way that a human being

would. Uh nothing super impressive on

the agentic coding aspect. GBT 5.6

actually still outperforms all of them.

So they were probably just deep trained

on deep s v1.1 on aentic coding. You can

see that it is essentially equal to

fable 5 although a big step up from opus

4.8. And what's cool is there's this new

automation bench which I'm particularly

interested in as somebody that does a

lot of automations for businesses. And

it scores 26% on that compared to 17.4%

for fable. And this is on building

workflows for businesses. So I mean

obviously it's a far cry away from

humans as of right now. You know

basically a quarter of the time it's

getting things right but you know it's a

big step up from just a couple of

generations ago. Legal is at 11.7%.

Health is at 59.8%

and then what's interesting is biology

it scores 49.4% on hard problems but

90.1% on human solved problems which is

really really interesting to see. One of

my favorite breakdowns to date is the

Aenta computers performance by effort

level. Essentially, they will feed

models a bunch of different tasks and

then they will see how many tokens and

uh you know, conversely how much money

they needed to spend in order to

complete those tasks. And so, you can

see over the course of uh you know, a

variety of effort levels from like super

low effort all the way up to high

there's quite the range and most of this

range is actually caused by GPD 5.6 soul

um which has a massive sort of spread of

cost per tasks and scores on their

lowest reasoning level all the way up to

their highest. Well, anyway, cloud

models tend to cluster sort of at the

top right hand corner of this graph,

which means there's less of a spread and

uh also less of a a task cost spread as

well. So, you know, if you think about

it, the dumbest version of Opus 5 still

scores about 60% on the vast majority of

agent computer use tasks um for a cost

of I want to say about $9 or so. That's

probably what that looks like. uh you

know, as you progress up the reasoning

ladder, the most intelligent models

score something like 70% for about $22 a

task, which is quite impressive,

especially considering Fable 5, you

know, can score approximately the same,

I want to say, as the mid-level of Opus

5, but it does so at nearly $50 as well.

So, massive cost saver there.

Performance on the new Automation Bench

is just leagues above virtually every

other model. As you can see, they tend

to cluster here around the I don't know

15% pass rate or so. This is almost 25%.

So big implications on people that want

to use models like this to automate

parts of their workflows. And I hate to

say it, but we're pretty close to

beating humanity's last exam. I mean,

you could see here even the dumbest

version of Opus 5 scores somewhere

around 56% or so. Uh, and the smartest

one is almost 65% across the board for a

cost per task of $3, which makes it not

only better at Fable, a better than

Fable at this particular benchmark, but

also marketkedly cheaper. And it's not

even really comparable to Opus 4.8.

There's also the ARC AGI 3 benchmark,

and on novel problem solving by cost, it

is just massive. Top right hand corner

here, you can see the vast majority of

the other models like Opus 4.8, GBD 5.6.

They didn't put Fable on this, I think,

just cuz they didn't run it on RAGI 3,

but they all scored somewhere around 2%.

And then over here, Opus 5 is just

brutally mogging them all at like 31 or

so. Finally, a mention about their

alignment. Um, misaligned behavior is

something that obviously large language

model companies like Enthropic really

want to crank down on. They hate when

models do things, you know, that are

sort of subversive or things that a

human being, you know, didn't really

necessarily want the model to do in

order to achieve the task. And as you

can see here, Opus 4.8 8 on this 10p

part score was a 2.85. Mythos 5, which

is that massive cyber security disaster

causer that uh led the US government to

pulling back a lot of this access was a

2.81. Sonnet 5 was at 3.35, but Opus 5

is way below all of them at 2.3, which

really just tells you, not only is this

model smarter, significantly cheaper,

significantly better at a lot of

specific tool calling tasks like the

automation bench and variety of terminal

bench workflows, but it's also a lot

safer, which really makes this a win-win

across the board. What does this mean

for AI more generally? Well, no real new

crazy unlocks here. I mean, Opus 5 is

continuing the slow and steady stream of

improvements that we saw with Opus 4,

Opus 4.5, Opus 4.7, Opus 4.8, and so on.

This sort of thing comes at a good time

because Kimmy K3 obviously dropped, you

know, about think about a week, a week

and a bit ago, and that really upset the

quote unquote balance of power between,

you know, these open-source or Chinese

models and then the more closed source

now American frontier models. I think as

we continue to go into the future,

what'll probably happen is, you know,

these models are getting quite

intelligent. They're getting quite ready

and capable of disrupting human

knowledge work at scale because of the

economic implications, I would imagine

that large AI model companies like

Anthropic, OpenAI, you know, now they're

colluding with the US government

essentially. I think what'll probably

happen is we're just going to see a slow

and steady drip release of these models.

But the actual intelligences, like the

really big galaxy brain ones, the ones

that are far smarter than anything we

saw in this benchmark, I think those are

going to repeatedly and more brazenly

just stay behind closed doors. And

unfortunately, I don't think us uh poor

little normies are going to have access

to them anytime soon. So do with that

what you will, but uh yeah, hopefully

now you know know everything you need in

order to crush it with Opus 5 integrated

into your business workflows and more.

If you guys like this video, definitely

check out Maker School down below. It's

my 90-day automation accountability

roadmap where I will guarantee you your

first customer selling something built

by Opus 5 or a similar model in 90 days

or I'll give you all your money back.

Have a lovely rest of the