Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

118 trillion tokens in one week, visualized as a crowd. I made a site that turned real OpenRouter data into a tiny town where every AI lab is a shop. More tokens, bigger crowd.

Ryan Sael

2,713 subscribers

141,561 görüntüleme • 7 gün önce •via X (Twitter)

Sanat Oyun Bilim & Teknoloji

Anya Rossi• Live Now

Private livecam show

45 Yorum

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Related project:

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

just added realtime visitor count, since this thing is getting viral and we're getting a thousand visitor in less than 12 hours!

imoutoftokens profil fotoğrafı

imoutoftokens7 gün önce

would love to scrub through previous weeks and watch the crowds change after a model release. such a nice way to make a leaderboard feel understandable

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

good idea but would also take a lot of data scrapping and to keep usage history, not sure if @OpenRouter would be happy with that

imoutoftokens profil fotoğrafı

imoutoftokens7 gün önce

@OpenRouter fair point, i was picturing the animation and skipping the data headache 😅 could be a lovely collaboration if they're up for it

Loktar 🇺🇸 profil fotoğrafı

Loktar 🇺🇸7 gün önce

Love this! Just the art style itself is cool.

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Thanks Loktar!

it8.art profil fotoğrafı

it8.art6 gün önce

Goodbye, GPT. I’ve decided to spend the $200 on DeepSeek-V41-Flash.

Ryan Sael profil fotoğrafı

Ryan Sael5 gün önce

Is it worth it? How much of a difference in performance?

it8.art profil fotoğrafı

it8.art5 gün önce

The difference isn’t that big. It’s good enough.

Ryan Sael profil fotoğrafı

Ryan Sael5 gün önce

right. I can't settle with good enough. I need the best in class performance. unless it's actually like 10x times more cost efficient. chatgpt subs actually still gives a lot more than paying direct API even with deepseek, maybe I'm wrong

Ferus profil fotoğrafı

Ferus7 gün önce

It gets you more hyped than cocaine

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

simple fact

man ki Hui profil fotoğrafı

man ki Hui6 gün önce

Awesome 🤩

Ryan Sael profil fotoğrafı

Ryan Sael6 gün önce

Thanks man ki Hui!

Chrono profil fotoğrafı

Chrono7 gün önce

love it!

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Thanks Chrono! which part you like the most?

Toven profil fotoğrafı

Toven1 gün önce

woah holy shit how did i miss this

Ryan Sael profil fotoğrafı

Ryan Sael1 gün önce

i had about 500 followers when I posted this

East Router profil fotoğrafı

East Router6 gün önce

This is super cool

Coder Schmoder profil fotoğrafı

Coder Schmoder6 gün önce

This is probably the coolest data visualization project that I have ever seen.

Ryan Sael profil fotoğrafı

Ryan Sael5 gün önce

Thanks man

-_- profil fotoğrafı

-_-6 gün önce

Amazing thought!

Wawa Sensei profil fotoğrafı

Wawa Sensei7 gün önce

Awesome 🤩 Could you reuse them for other use cases? Could be a super motion graphics tool!

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Thanks! yes very possible to reuse the effect

Dawson Russell profil fotoğrafı

Dawson Russell7 gün önce

love this.

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Thanks Dawson!

Abdullah Saleh profil fotoğrafı

Abdullah Saleh6 gün önce

Perfect idea

Dzmitry profil fotoğrafı

Dzmitry7 gün önce

Love the UI and visualization!

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

Thanks Dzmitry! Which part do you think most stand out?

J A Z I I profil fotoğrafı

J A Z I I6 gün önce

Wow crazy

toodles891 profil fotoğrafı

toodles8917 gün önce

So sick!

Hari profil fotoğrafı

Hari7 gün önce

Astra or fable?

Ryan Sael profil fotoğrafı

Ryan Sael7 gün önce

is coding a sin these days?

Hari profil fotoğrafı

Hari6 gün önce

Willingly wasting the most precious currency (time) we humans have is sin. Yes.

Andy🤖 profil fotoğrafı

Andy🤖6 gün önce

so cool, always had the same idea in my head about turning learning process or business into a game, i mean, it makes you control almost every detail of it, and in the same time, its super interesting just to sit back and observing it, let it unfold. Great work, Ryan!!!

AI Mastery Guide profil fotoğrafı

AI Mastery Guide7 gün önce

turning tokens into a whole town, cool

Bob Muțe profil fotoğrafı

Bob Muțe6 gün önce

Brilliant

Shohan profil fotoğrafı

Shohan7 gün önce

coool

Daniel Neuhaus profil fotoğrafı

Daniel Neuhaus7 gün önce

What do the people represent? Agents or tokens or tasks or what?

Dark Star profil fotoğrafı

Dark Star7 gün önce

try this

Synaptic labs profil fotoğrafı

Synaptic labs4 gün önce

Coming soon 🔜

Stackaible profil fotoğrafı

Stackaible6 gün önce

Brilliant!

Matthis Linck profil fotoğrafı

Matthis Linck7 gün önce

Love the look an feel of it

Maheshwaran Umapathy profil fotoğrafı

Maheshwaran Umapathy1 gün önce

What's the prompt and harness used?

Benzer Videolar

This person finding out V that steps into the crowd is actually real, looked him up and now wants to be a fan, im screamingg V stepped into the crowd and turned everyone into a fan

This person finding out V that steps into the crowd is actually real, looked him up and now wants to be a fan, im screamingg V stepped into the crowd and turned everyone into a fan

K⁷ | 13TS

65,752 görüntüleme • 1 ay önce

A viral tweet claimed that Meta is now responsible for nearly a third of Anthropic's $30B ARR, based on reporting that Meta used 60.2T tokens over 30 days. This doesn't actually pencil out. It assumes all of Meta's spend is on output tokens, which are much more expensive than input tokens, and which would be extremely unusual — according to OpenRouter data, about 98.9% of all Opus 4.6 tokens are input. The actual number is *maybe* closer to $136M a month, or $1.6B a year. Tyler Cosgrove goes through the math:

A viral tweet claimed that Meta is now responsible for nearly a third of Anthropic's $30B ARR, based on reporting that Meta used 60.2T tokens over 30 days. This doesn't actually pencil out. It assumes all of Meta's spend is on output tokens, which are much more expensive than input tokens, and which would be extremely unusual — according to OpenRouter data, about 98.9% of all Opus 4.6 tokens are input. The actual number is maybe closer to $136M a month, or $1.6B a year. Tyler Cosgrove goes through the math:

TBPN

254,894 görüntüleme • 5 ay önce

At China’s largest wholesale market, customers crowd one small shop looking for the so called ‘crying horse,’ a red plush toy meant as a Lunar New Year decoration that has gone viral after a manufacturing mistake turned its smile into a frown

At China’s largest wholesale market, customers crowd one small shop looking for the so called ‘crying horse,’ a red plush toy meant as a Lunar New Year decoration that has gone viral after a manufacturing mistake turned its smile into a frown

Reuters

60,158 görüntüleme • 8 ay önce

$A house, a film and a forest all have one thing in common. They are real world assets that can be represented as tokens or fractionalized tokens on a blockchain. Ripple’s Jas Takhar describes the potential of a tokenized future in under 60 seconds. ⏱️ #CryptoInOneMinute$

A house, a film and a forest all have one thing in common. They are real world assets that can be represented as tokens or fractionalized tokens on a blockchain. Ripple’s Jas Takhar describes the potential of a tokenized future in under 60 seconds. ⏱️ #CryptoInOneMinute

Ripple

181,820 görüntüleme • 3 yıl önce

10x more tokens/second per megawatt as measured by CoreWeave on NVIDIA Vera Rubin platform. That performance reflects a bigger shift in AI infrastructure economics: building AI factories that efficiently turn compute into tokens, and tokens into revenue. It takes extreme co-design across the full stack, combined with an open ecosystem built to innovate and scale. Dion Harris connects the economics to the technology with 5 critical questions every AI infrastructure leader should be asking across compute, networking, storage, software and security. ➡️

10x more tokens/second per megawatt as measured by CoreWeave on NVIDIA Vera Rubin platform. That performance reflects a bigger shift in AI infrastructure economics: building AI factories that efficiently turn compute into tokens, and tokens into revenue. It takes extreme co-design across the full stack, combined with an open ecosystem built to innovate and scale. Dion Harris connects the economics to the technology with 5 critical questions every AI infrastructure leader should be asking across compute, networking, storage, software and security. ➡️

NVIDIA AI Infrastructure

23,895 görüntüleme • 1 ay önce

Managing your tokens with Family is as easy as drag-and-drop. Hide, star or trash tokens one at a time or in bulk, with intuitive gestures and delightful interactions at every stage. It's really that simple.🤌

Managing your tokens with Family is as easy as drag-and-drop. Hide, star or trash tokens one at a time or in bulk, with intuitive gestures and delightful interactions at every stage. It's really that simple.🤌

Family

55,499 görüntüleme • 3 yıl önce

“The currency for AI is crypto.” ~CZ Humans generate 300 trillion tokens a day. Soon, AI will generate 10× that. As AI begins trading with AI, and blockchain secures data and payments, productivity decouples from population. CZ predicts GDP may no longer be tied to people.

“The currency for AI is crypto.” ~CZ Humans generate 300 trillion tokens a day. Soon, AI will generate 10× that. As AI begins trading with AI, and blockchain secures data and payments, productivity decouples from population. CZ predicts GDP may no longer be tied to people.

vitrupo

261,267 görüntüleme • 1 yıl önce

.Jay Town to Jack Posobiec on the Tyler Robinson hearing: "Because of the tinfoil hat crowd that has taken nothingburgers and turned it into a smogasbord, I think it's actually a really good idea in this particular case to have as much public as possible."

.Jay Town to Jack Posobiec on the Tyler Robinson hearing: "Because of the tinfoil hat crowd that has taken nothingburgers and turned it into a smogasbord, I think it's actually a really good idea in this particular case to have as much public as possible."

Human Events

15,133 görüntüleme • 2 ay önce

Police have just made a VIOLENT arrest in LIVERPOOL where several officers pounced on one man. The crowd is riled up after witnessing that. Understandably.

Police have just made a VIOLENT arrest in LIVERPOOL where several officers pounced on one man. The crowd is riled up after witnessing that. Understandably.

The Nationalist GB

30,374 görüntüleme • 3 ay önce

ONE-HANDED FAN GRAB 🤯 Kane Williamson smokes one into the crowd where a stunning catch is taken 🚨

ONE-HANDED FAN GRAB 🤯 Kane Williamson smokes one into the crowd where a stunning catch is taken 🚨

Sky Sports Cricket

121,863 görüntüleme • 1 yıl önce

Every Currency on earth is now available as a pairable commodity for your tokens. Real World Currency Markets are now live.

Every Currency on earth is now available as a pairable commodity for your tokens. Real World Currency Markets are now live.

Commodity Market Exchange

31,082 görüntüleme • 12 gün önce

🚨NEW JERSEY Trump-Endorsed candidate for Governor Jack Ciattarelli is having a town hall in Essex County that is PACKED OUT THE DOOR, the crowd spilling into the street This is in the BLUEST county in the state, a crowd like this for a Republican here is unheard of

🚨NEW JERSEY Trump-Endorsed candidate for Governor Jack Ciattarelli is having a town hall in Essex County that is PACKED OUT THE DOOR, the crowd spilling into the street This is in the BLUEST county in the state, a crowd like this for a Republican here is unheard of

Michael Casey

713,358 görüntüleme • 11 ay önce

If I could invest into one private company right now it would be Fireworks Fireworks AI serves 40 trillion tokens a day. They run inference for open-weight and customer-tuned models. Over 95% of those tokens come from models tuned on a customer's own data rather than off-the-shelf frontier models. Revenue passed $1B annualized in July, up 5x year over year. They raised $1.505B at a $17.5B valuation on July 15. Eight months earlier they were valued at $4B. "Fireworks is the TSMC of AI Factories." Jensen Huang

If I could invest into one private company right now it would be Fireworks Fireworks AI serves 40 trillion tokens a day. They run inference for open-weight and customer-tuned models. Over 95% of those tokens come from models tuned on a customer's own data rather than off-the-shelf frontier models. Revenue passed $1B annualized in July, up 5x year over year. They raised $1.505B at a $17.5B valuation on July 15. Eight months earlier they were valued at $4B. "Fireworks is the TSMC of AI Factories." Jensen Huang

KawzInvests

19,940 görüntüleme • 1 ay önce

MiniMax M3 turned a napkin sketch into a playable game We handed MiniMax M3 a hand-drawn draft of a Doodle Jump style platformer. It read the elements off the draft, wrote the logic, drew the interface and shipped it as one self-contained HTML game Input: 6,920 tokens Output: 9,933 tokens Cost: $0.028 MiniMax (official) drops M3 on Hugging Face next week

MiniMax M3 turned a napkin sketch into a playable game We handed MiniMax M3 a hand-drawn draft of a Doodle Jump style platformer. It read the elements off the draft, wrote the logic, drew the interface and shipped it as one self-contained HTML game Input: 6,920 tokens Output: 9,933 tokens Cost: $0.028 MiniMax (official) drops M3 on Hugging Face next week

atomic.chat

64,640 görüntüleme • 3 ay önce

$AI shouldn’t be controlled by a single company or person, and the reason why most AI tools are "free" is because you are the product. Your data is more valuable to them, than the tool itself. That's why Action Model is different. We use tokens so: • Every AI action has a real cost • That cost is transparent • Value flows back to contributors, not one corporation The token represents a fractional stake in the AI itself; economic rights, governance rights, and participation in its growth. And that’s how you run AI without centralized control.$

AI shouldn’t be controlled by a single company or person, and the reason why most AI tools are "free" is because you are the product. Your data is more valuable to them, than the tool itself. That's why Action Model is different. We use tokens so: • Every AI action has a real cost • That cost is transparent • Value flows back to contributors, not one corporation The token represents a fractional stake in the AI itself; economic rights, governance rights, and participation in its growth. And that’s how you run AI without centralized control.

Action Model

47,564 görüntüleme • 7 ay önce

New video of Qualcomm CEO Cristiano Amon: AI will require “gazillions” of tokens. Because, Agentic AI will consume dramatically more tokens because it performs autonomous tasks, uses multiple systems, and interacts with tools. AI demand will grow hugely when software starts letting agents act, not just answer. A chatbot spends tokens on language; an agent spends tokens on deciding, checking, calling tools, reading outputs, revising plans, and coordinating with other software. Today a single human-AI exchange may be large, a reasoning task may be much larger, but we are already entering the agentic era, where an autonomous workflow can become exponentially larger still because the model is no longer producing one response. It is running a process. When a SaaS product redesigns itself around human-agent interaction, every task can become a chain of hidden micro-decisions, and each micro-decision consumes context, memory, tool calls, verification, and output tokens. Another point is people hear “more tokens” and think only of bigger data centers, when the deeper change is economic: software usage may stop being measured mainly by clicks, seats, or sessions. It will be measured by how much machine reasoning/tokens gets spent on behalf of each user. ---- From "Reuters" YouTube channel, (link in comment)

New video of Qualcomm CEO Cristiano Amon: AI will require “gazillions” of tokens. Because, Agentic AI will consume dramatically more tokens because it performs autonomous tasks, uses multiple systems, and interacts with tools. AI demand will grow hugely when software starts letting agents act, not just answer. A chatbot spends tokens on language; an agent spends tokens on deciding, checking, calling tools, reading outputs, revising plans, and coordinating with other software. Today a single human-AI exchange may be large, a reasoning task may be much larger, but we are already entering the agentic era, where an autonomous workflow can become exponentially larger still because the model is no longer producing one response. It is running a process. When a SaaS product redesigns itself around human-agent interaction, every task can become a chain of hidden micro-decisions, and each micro-decision consumes context, memory, tool calls, verification, and output tokens. Another point is people hear “more tokens” and think only of bigger data centers, when the deeper change is economic: software usage may stop being measured mainly by clicks, seats, or sessions. It will be measured by how much machine reasoning/tokens gets spent on behalf of each user. ---- From "Reuters" YouTube channel, (link in comment)

Rohan Paul

31,617 görüntüleme • 3 ay önce

AI will require “gazillions” of tokens. - Qualcomm CEO Cristiano Amon: Agentic AI will consume dramatically more tokens because it performs autonomous tasks, uses multiple systems, and interacts with tools. A chatbot spends tokens on language; an agent spends tokens on deciding, checking, calling tools, reading outputs, revising plans, and coordinating with other software. So, AI demand will grow hugely when software starts letting agents act, not just answer. Today a single human-AI exchange may be large, a reasoning task may be much larger, but we are already entering the agentic era, where an autonomous workflow can become exponentially larger still because the model is no longer producing one response. It is running a process. When a SaaS product redesigns itself around human-agent interaction, every task can become a chain of hidden micro-decisions, and each micro-decision consumes context, memory, tool calls, verification, and output tokens. Another point is people hear “more tokens” and think only of bigger data centers, when the deeper change is economic: software usage may stop being measured mainly by clicks, seats, or sessions. It will be measured by how much machine reasoning/tokens gets spent on behalf of each user. ---- From "Reuters" YouTube channel, (link in comment)

AI will require “gazillions” of tokens. - Qualcomm CEO Cristiano Amon: Agentic AI will consume dramatically more tokens because it performs autonomous tasks, uses multiple systems, and interacts with tools. A chatbot spends tokens on language; an agent spends tokens on deciding, checking, calling tools, reading outputs, revising plans, and coordinating with other software. So, AI demand will grow hugely when software starts letting agents act, not just answer. Today a single human-AI exchange may be large, a reasoning task may be much larger, but we are already entering the agentic era, where an autonomous workflow can become exponentially larger still because the model is no longer producing one response. It is running a process. When a SaaS product redesigns itself around human-agent interaction, every task can become a chain of hidden micro-decisions, and each micro-decision consumes context, memory, tool calls, verification, and output tokens. Another point is people hear “more tokens” and think only of bigger data centers, when the deeper change is economic: software usage may stop being measured mainly by clicks, seats, or sessions. It will be measured by how much machine reasoning/tokens gets spent on behalf of each user. ---- From "Reuters" YouTube channel, (link in comment)

Rohan Paul

12,664 görüntüleme • 2 ay önce

He walked into a small shop and made it visible to the entire planet. One afternoon. 170 scans. A free app on his phone. He gets $400 for the scan. $99 every month for hosting. The result: a full 3D virtual tour anyone can walk through in a browser. Every shelf. Every product. Every detail of a store that most people would never find on a map. The owner gets more online visitors in a week than she got in-store visitors in a month. That gap is the business. Month one: $3,500. Month six: $18,000. The same model works for hotels, real estate listings, event venues, car dealerships, museums. Every physical space in every city is an untapped asset. Most owners have no idea this exists. He didn't wait for the right client. He walked into a shop and made one.

He walked into a small shop and made it visible to the entire planet. One afternoon. 170 scans. A free app on his phone. He gets $400 for the scan. $99 every month for hosting. The result: a full 3D virtual tour anyone can walk through in a browser. Every shelf. Every product. Every detail of a store that most people would never find on a map. The owner gets more online visitors in a week than she got in-store visitors in a month. That gap is the business. Month one: $3,500. Month six: $18,000. The same model works for hotels, real estate listings, event venues, car dealerships, museums. Every physical space in every city is an untapped asset. Most owners have no idea this exists. He didn't wait for the right client. He walked into a shop and made one.

Superior

16,952 görüntüleme • 4 ay önce

Gavin Baker Gavin Baker: the Anthropic S1 is going to break a lot of people's brains. The entire AI bear case rests on one assumption, that tokens are sold below cost. What the filing settles, and who has to cover: Baker says that assumption is wrong. Open-source tokens are profitable. Anthropic is profitable and generating cash. OpenAI is close. Most tokens in the chain make money today. The S1 is the first audited document that settles it either way. Jim Chanos James Chanos has spent the year on the other side, calling the AI buildout a circular financing story. Plain version for a position holder: a lot of macro and value money is bearish on AI because it believes tokens are subsidized. If the filing shows real margins, that money covers into a re-rating and $NVDA, $TSM, $MU and the hyperscalers all move on it. If the filing shows subsidized tokens, every valuation downstream of it is sitting on sand. One document, one binary, a date on the calendar. David Sacks David Sacks and Jason Calacanis @jason worked him over on it for a full segment. Source: All-In Podcast -

Gavin Baker Gavin Baker: the Anthropic S1 is going to break a lot of people's brains. The entire AI bear case rests on one assumption, that tokens are sold below cost. What the filing settles, and who has to cover: Baker says that assumption is wrong. Open-source tokens are profitable. Anthropic is profitable and generating cash. OpenAI is close. Most tokens in the chain make money today. The S1 is the first audited document that settles it either way. Jim Chanos James Chanos has spent the year on the other side, calling the AI buildout a circular financing story. Plain version for a position holder: a lot of macro and value money is bearish on AI because it believes tokens are subsidized. If the filing shows real margins, that money covers into a re-rating and $NVDA, $TSM, $MU and the hyperscalers all move on it. If the filing shows subsidized tokens, every valuation downstream of it is sitting on sand. One document, one binary, a date on the calendar. David Sacks David Sacks and Jason Calacanis @jason worked him over on it for a full segment. Source: All-In Podcast -

Podcast Alpha

29,416 görüntüleme • 1 ay önce