Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Gavin Baker: "Anthropic is roughly four times more capital efficient than OpenAI - they're growing really fast and they're not burning that much cash" this is him explaining why he thinks token efficiency decides who wins, why xAI has the lowest cost per token of anyone, and what a...

carm1nee's profile picture

Сarm1ne

15,013 subscribers

128,347 Aufrufe • vor 1 Monat •via X (Twitter)

Wissenschaft & TechnologieBildung
LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

$NVDA has the “lowest cost per token in the world” which is really a claim about owning the economics of inference. The company that delivers the most tokens per watt, per rack and per dollar is the one that captures the most demand and the best margins as AI scales.
0:47

$NVDA has the “lowest cost per token in the world” which is really a claim about owning the economics of inference. The company that delivers the most tokens per watt, per rack and per dollar is the one that captures the most demand and the best margins as AI scales.

Shay Boloor

48,594 Aufrufe • vor 5 Monaten

“NVIDIA’s cost per token is the lowest in the world.” — Jensen Huang, Founder & CEO of NVIDIA Token generation cost is a direct result of architecture excellence and extreme co-design, not just compute cost. Lowest cost per token and highest performance per watt are definitive measures of AI economics and the key to unlocking maximum profitability and AI revenue. ➡️
0:49

“NVIDIA’s cost per token is the lowest in the world.” — Jensen Huang, Founder & CEO of NVIDIA Token generation cost is a direct result of architecture excellence and extreme co-design, not just compute cost. Lowest cost per token and highest performance per watt are definitive measures of AI economics and the key to unlocking maximum profitability and AI revenue. ➡️

NVIDIA

89,074 Aufrufe • vor 5 Monaten

🎙️ Want a full breakdown on why cost per token is a key metric for AI infrastructure TCO? We’re operating in a power-constrained world. Demand for intelligence is scaling faster than infrastructure can keep up—so metrics like cost per GPU hour or FLOPs per dollar don’t tell the full story. When it comes to TCO, Cost per Token is the metric that captures end-to-end performance across GPUs, CPUs, networking, software, and ecosystem making it the key driver of real profitability and scalability in AI. If you’re not measuring efficiency at the token level, you’re missing what actually drives cost. Watch the deep-dive conversation with Dr. Gerro Prinsloo , Nader Khalil and Carter Abdallah, live on the NVIDIA YouTube channel.
1:05

🎙️ Want a full breakdown on why cost per token is a key metric for AI infrastructure TCO? We’re operating in a power-constrained world. Demand for intelligence is scaling faster than infrastructure can keep up—so metrics like cost per GPU hour or FLOPs per dollar don’t tell the full story. When it comes to TCO, Cost per Token is the metric that captures end-to-end performance across GPUs, CPUs, networking, software, and ecosystem making it the key driver of real profitability and scalability in AI. If you’re not measuring efficiency at the token level, you’re missing what actually drives cost. Watch the deep-dive conversation with Dr. Gerro Prinsloo , Nader Khalil and Carter Abdallah, live on the NVIDIA YouTube channel.

NVIDIA

21,970 Aufrufe • vor 5 Monaten

"SpaceX AI is doing something that Anthropic and Open AI both didn't and couldn't do... what is xAI doing? xAI is building data centers, right now on Earth, that are larger, more productive, and they're built faster and at lower cost than any other data centers in the world... so when you look at that and you actually plot that out, 1.8 trillion may be low." Larry Goldberg
4:33

"SpaceX AI is doing something that Anthropic and Open AI both didn't and couldn't do... what is xAI doing? xAI is building data centers, right now on Earth, that are larger, more productive, and they're built faster and at lower cost than any other data centers in the world... so when you look at that and you actually plot that out, 1.8 trillion may be low." Larry Goldberg

Ellie in Space 🚀💫

58,309 Aufrufe • vor 3 Monaten

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.
0:42

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 Aufrufe • vor 7 Monaten

"OpenAI and Anthropic are pretty different animals from a capital efficiency perspective. And Anthropic clearly has a dramatically lower cost per token than OpenAI. They burned maybe 80% less than OpenAI. So as businesses, they clearly have very different structural ROICs." - A solid, insightful interview featuring Gavin Baker (Gavin Baker), U.S. hedge fund manager and Chief Investment Officer of Atreides Management. ---- From "Invest Like The Best" (Invest Like the Best) YouTube channel, (link in comment)
1:12

"OpenAI and Anthropic are pretty different animals from a capital efficiency perspective. And Anthropic clearly has a dramatically lower cost per token than OpenAI. They burned maybe 80% less than OpenAI. So as businesses, they clearly have very different structural ROICs." - A solid, insightful interview featuring Gavin Baker (Gavin Baker), U.S. hedge fund manager and Chief Investment Officer of Atreides Management. ---- From "Invest Like The Best" (Invest Like the Best) YouTube channel, (link in comment)

Rohan Paul

25,185 Aufrufe • vor 2 Monaten

Gavin on how Blackwell and GB300 will end Google's position as the lowest-cost producer of tokens: "I think it's important that Google has been the lowest cost producer of tokens. And this is really important because AI is the first time in my career as a tech investor that being the low cost producer has ever mattered. What Google has been doing as the low cost producer is they have been sucking the economic oxygen out of the AI ecosystem, which is an extremely rational strategy for them -- let's make life really hard for our competitors. We'll see the first models trained on Blackwell in early 2026. I think the first Blackwell model will come from xAI because they build the data centers the fastest. The GB300 is a great chip. It is drop-in compatible in every way with those GB200 racks, and the companies that use the GB300s -- they're going to be the low cost producer of tokens. I think this has pretty profound implications because it has to change Google's strategic calculus. If you have decisive cost advantage, why not AI at a -30% margin. You make it hard for your competitors who need funding to raise the capital that they need. On the other side of that maybe you have an extremely dominant share position. Well, all of that changes once Google is no longer the low-cost producer."
3:33

Gavin on how Blackwell and GB300 will end Google's position as the lowest-cost producer of tokens: "I think it's important that Google has been the lowest cost producer of tokens. And this is really important because AI is the first time in my career as a tech investor that being the low cost producer has ever mattered. What Google has been doing as the low cost producer is they have been sucking the economic oxygen out of the AI ecosystem, which is an extremely rational strategy for them -- let's make life really hard for our competitors. We'll see the first models trained on Blackwell in early 2026. I think the first Blackwell model will come from xAI because they build the data centers the fastest. The GB300 is a great chip. It is drop-in compatible in every way with those GB200 racks, and the companies that use the GB300s -- they're going to be the low cost producer of tokens. I think this has pretty profound implications because it has to change Google's strategic calculus. If you have decisive cost advantage, why not AI at a -30% margin. You make it hard for your competitors who need funding to raise the capital that they need. On the other side of that maybe you have an extremely dominant share position. Well, all of that changes once Google is no longer the low-cost producer."

Patrick OShaughnessy

174,768 Aufrufe • vor 9 Monaten

OpenAI chairman Bret Taylor talks to about 100 CEOs every month. His answer to the cheap open-weight model panic: cheaper to train does not mean cheaper to use, and the number that decides it is token efficiency. "One thing that I think is a little bit overblown about these open weight models is they're not necessarily cheaper to run. Whether or not they're cheaper to train, you don't care. Because you're using just as many tokens. In fact, they may be less efficient." "There's this thing called token efficiency. And it turns out the frontier models are much, much more token efficient." "A token is to intelligence like a watt is to electricity... how many tokens does it take to complete a task? Not every token is actually equal." "For a lot of tasks, it turns out these frontier models from OpenAI and Anthropic are actually just better than these open weight models... just having open weights isn't actually the main thing driving any of those costs." Later in the same interview he goes after the billing unit itself: "It would be like if you signed up for Gmail and you paid for CPU cycle or something... where the world is going is paying for outcomes." The unresolved column: the chart CNBC airs mid-answer, from Artificial Analysis, prices a completed task at $0.94 on Kimi K3 against $2.75 on Claude Fable 5, efficiency folded in. If that gap holds, the premium he is defending gets earned on quality, not price. - Bret Taylor (Bret Taylor), OpenAI chairman and Sierra co-founder, on CNBC's Squawk Box.
2:09

OpenAI chairman Bret Taylor talks to about 100 CEOs every month. His answer to the cheap open-weight model panic: cheaper to train does not mean cheaper to use, and the number that decides it is token efficiency. "One thing that I think is a little bit overblown about these open weight models is they're not necessarily cheaper to run. Whether or not they're cheaper to train, you don't care. Because you're using just as many tokens. In fact, they may be less efficient." "There's this thing called token efficiency. And it turns out the frontier models are much, much more token efficient." "A token is to intelligence like a watt is to electricity... how many tokens does it take to complete a task? Not every token is actually equal." "For a lot of tasks, it turns out these frontier models from OpenAI and Anthropic are actually just better than these open weight models... just having open weights isn't actually the main thing driving any of those costs." Later in the same interview he goes after the billing unit itself: "It would be like if you signed up for Gmail and you paid for CPU cycle or something... where the world is going is paying for outcomes." The unresolved column: the chart CNBC airs mid-answer, from Artificial Analysis, prices a completed task at $0.94 on Kimi K3 against $2.75 on Claude Fable 5, efficiency folded in. If that gap holds, the premium he is defending gets earned on quality, not price. - Bret Taylor (Bret Taylor), OpenAI chairman and Sierra co-founder, on CNBC's Squawk Box.

Karl Mehta

17,254 Aufrufe • vor 1 Monat

Chamath Palihapitiya on the Tesla-SpaceX merger: "I still don't like the fact that Tesla's over here - and as I've told you, that will get merged in. that thing will look very cheap in a few years" this is him and Gavin Baker explaining how you underwrite SpaceX at $2 trillion, why it builds data centers faster than anyone alive, and what happened when Elon let a company nobody believed in onto Colossus "their first data center took 122 days. the second took 91. the third was 66 - they build data centers dramatically faster than anyone else at a lower cost, and GPUs get allocated to whoever can plug them in, turn them on, and start converting electrons into tokens" "Anthropic is paying $1.25 billion a month to rent Colossus - in other words they added a Starlink in terms of revenue" "Cursor was dead in the water on compute, falling far behind Codex, Google and Anthropic - then Elon let them on Colossus, and three weeks later their model was Pareto dominant" "last year it did 18, 19 billion, it'll probably do 25 to 30 this year, next year 40 to 45, and the year after that it doubles again - so now I'm buying it at 20 times revenue" bookmark & watch the full conversation ↓
12:15

Chamath Palihapitiya on the Tesla-SpaceX merger: "I still don't like the fact that Tesla's over here - and as I've told you, that will get merged in. that thing will look very cheap in a few years" this is him and Gavin Baker explaining how you underwrite SpaceX at $2 trillion, why it builds data centers faster than anyone alive, and what happened when Elon let a company nobody believed in onto Colossus "their first data center took 122 days. the second took 91. the third was 66 - they build data centers dramatically faster than anyone else at a lower cost, and GPUs get allocated to whoever can plug them in, turn them on, and start converting electrons into tokens" "Anthropic is paying $1.25 billion a month to rent Colossus - in other words they added a Starlink in terms of revenue" "Cursor was dead in the water on compute, falling far behind Codex, Google and Anthropic - then Elon let them on Colossus, and three weeks later their model was Pareto dominant" "last year it did 18, 19 billion, it'll probably do 25 to 30 this year, next year 40 to 45, and the year after that it doubles again - so now I'm buying it at 20 times revenue" bookmark & watch the full conversation ↓

Сarm1ne

176,651 Aufrufe • vor 1 Monat

Gavin Baker: "which AI company actually survives? Google - already multi-trillion Microsoft, OpenAI, Meta chose open source - the only ones left is xAI and Anthropic - xAI + X will be worth over a trillion" this is him explaining what he actually invests in - and why Taiwan sells sand to Nvidia for $700 that Nvidia sells for $50,000 His words: what to invest in: "we're investing in things that increase the GPU utilization rate - because if you take a GPU from 30% utilized to 60%, you've doubled the output of that AI factory" what not to Invest in: "the models will basically go to zero - the data center is a commodity - energy might be short for a while but is also a commodity - the only thing in the whole stack with real value is the data moat and reinforcement learning" "the GPU has gotten 50 times faster in the last four years - that's a big part of why we're living through this AI revolution" he doesn't rent intelligence - he owns the thing it runs on bookmark & watch - see how below with local agent ↓
13:10

Gavin Baker: "which AI company actually survives? Google - already multi-trillion Microsoft, OpenAI, Meta chose open source - the only ones left is xAI and Anthropic - xAI + X will be worth over a trillion" this is him explaining what he actually invests in - and why Taiwan sells sand to Nvidia for $700 that Nvidia sells for $50,000 His words: what to invest in: "we're investing in things that increase the GPU utilization rate - because if you take a GPU from 30% utilized to 60%, you've doubled the output of that AI factory" what not to Invest in: "the models will basically go to zero - the data center is a commodity - energy might be short for a while but is also a commodity - the only thing in the whole stack with real value is the data moat and reinforcement learning" "the GPU has gotten 50 times faster in the last four years - that's a big part of why we're living through this AI revolution" he doesn't rent intelligence - he owns the thing it runs on bookmark & watch - see how below with local agent ↓

bodila

312,169 Aufrufe • vor 1 Monat

⚡️ JUST IN: MUSK LAUNCHES GROK 4.5, CALLING IT "OPUS-CLASS" AI MODEL Musk just released Grok 4.5 on Grok Build, Cursor, and the SpaceXAI Console, calling it an "Opus-class" AI model. He says it is faster, more token-efficient and lower cost, with EU access expected by mid-July. Built on a 1.5 TRILLION parameter foundation model, it drops the same day as OpenAI's GPT-5.6. API pricing is $2 per million input tokens and $6 per million output tokens.
0:55

⚡️ JUST IN: MUSK LAUNCHES GROK 4.5, CALLING IT "OPUS-CLASS" AI MODEL Musk just released Grok 4.5 on Grok Build, Cursor, and the SpaceXAI Console, calling it an "Opus-class" AI model. He says it is faster, more token-efficient and lower cost, with EU access expected by mid-July. Built on a 1.5 TRILLION parameter foundation model, it drops the same day as OpenAI's GPT-5.6. API pricing is $2 per million input tokens and $6 per million output tokens.

Coin Bureau

50,488 Aufrufe • vor 2 Monaten

🚨 ELON MUSK SAYS YOU’LL SOON BE ABLE TO FALL ASLEEP IN A TESLA — AND WAKE UP AT YOUR DESTINATION. He calls Full Self-Driving “an AI-driven car.” “Tesla AI software that drives the car just by looking like a human does.” He says Tesla has “the most advanced real world AI” and Europe is about to see it. Approval in the Netherlands? Reportedly March 20. And then this: “This year, you’ll be able to fall asleep in the Tesla and wake up at your destination.” Not driver assist. Autonomous. Are we watching the iPhone moment for driving?
1:02

🚨 ELON MUSK SAYS YOU’LL SOON BE ABLE TO FALL ASLEEP IN A TESLA — AND WAKE UP AT YOUR DESTINATION. He calls Full Self-Driving “an AI-driven car.” “Tesla AI software that drives the car just by looking like a human does.” He says Tesla has “the most advanced real world AI” and Europe is about to see it. Approval in the Netherlands? Reportedly March 20. And then this: “This year, you’ll be able to fall asleep in the Tesla and wake up at your destination.” Not driver assist. Autonomous. Are we watching the iPhone moment for driving?

HustleBitch

24,619 Aufrufe • vor 6 Monaten

Jensen Huang gave Tesla one of the strongest endorsements possible: “I think that the Tesla stack is the most advanced AV stack in the world” “And I think that the Tesla AV operations is the most advanced in the world” “I’m fairly certain that they were already using end-to-end AI. And whether their AI also did reasoning or not is somewhat secondary to that first part” Tesla has a complete end-to-end AI stack powered by real-world data, massive training infrastructure, vertically integrated hardware and software, and continuous deployment across its fleet
0:26

Jensen Huang gave Tesla one of the strongest endorsements possible: “I think that the Tesla stack is the most advanced AV stack in the world” “And I think that the Tesla AV operations is the most advanced in the world” “I’m fairly certain that they were already using end-to-end AI. And whether their AI also did reasoning or not is somewhat secondary to that first part” Tesla has a complete end-to-end AI stack powered by real-world data, massive training infrastructure, vertically integrated hardware and software, and continuous deployment across its fleet

X Freeze

23,220 Aufrufe • vor 1 Monat

Ran Neuner shares his checklist for identifying protocols actually worth investing in "What are the protocols that are actually going to succeed? Number one, big daddy is Hyperliquid. That's why the market's giving Hyperliquid an $81 billion valuation. Why? Because they buy back tokens and burn them with fees every single day. Their fees in the last couple, has it been $4 million, $5 million a day? And they literally take that and buy back tokens with a large percentage of that money." "Pumpfun, same thing. Massive use case, growing. It's a network, definitely, because the more users that join the network, the more fun that network becomes. And they're buying back and burning the tokens. They tick all the boxes." " Venice, the AI token, they've got 4 million users, $100 million in annualized revenue, growing extremely fast, and a mechanism to give the value back to token holders." "So I think that's the formula. Are you growing? Are you growing into a network? Do you have real sustainable revenues? And do you have a mechanism to actually give that revenue back to your token holders? If the answer to all four of those is yes, invest." "If the answer to the first is yes and the fourth one is no, also invest."
1:10

Ran Neuner shares his checklist for identifying protocols actually worth investing in "What are the protocols that are actually going to succeed? Number one, big daddy is Hyperliquid. That's why the market's giving Hyperliquid an $81 billion valuation. Why? Because they buy back tokens and burn them with fees every single day. Their fees in the last couple, has it been $4 million, $5 million a day? And they literally take that and buy back tokens with a large percentage of that money." "Pumpfun, same thing. Massive use case, growing. It's a network, definitely, because the more users that join the network, the more fun that network becomes. And they're buying back and burning the tokens. They tick all the boxes." " Venice, the AI token, they've got 4 million users, $100 million in annualized revenue, growing extremely fast, and a mechanism to give the value back to token holders." "So I think that's the formula. Are you growing? Are you growing into a network? Do you have real sustainable revenues? And do you have a mechanism to actually give that revenue back to your token holders? If the answer to all four of those is yes, invest." "If the answer to the first is yes and the fourth one is no, also invest."

The Wolf Of All Streets

35,089 Aufrufe • vor 17 Tagen

Sam Altman on how AI token usage is growing at an exponential rate. "6 and a half years ago, the world token leader was an OpenAI employee using about 100,000 tokens a month. And this seemed ludicrous at the time. The worldwide average per capita was, like, zero. Now, the worldwide average of tokens per month, is like 100,000, and the token leader at OpenAI uses something in the hundreds of billions. If that happens again, which I think it probably will, then in another six and a half years, the average person uses, let’s say, 500 billion tokens a month, and the token leader uses a quadrillion or quadrillions of tokens a month. And I think that will just become the expectation." --- He was at YC Startup School 2026, in conversation Garry Tan From "Y Combinator" YouTube channel, (full video link in comment)
0:48

Sam Altman on how AI token usage is growing at an exponential rate. "6 and a half years ago, the world token leader was an OpenAI employee using about 100,000 tokens a month. And this seemed ludicrous at the time. The worldwide average per capita was, like, zero. Now, the worldwide average of tokens per month, is like 100,000, and the token leader at OpenAI uses something in the hundreds of billions. If that happens again, which I think it probably will, then in another six and a half years, the average person uses, let’s say, 500 billion tokens a month, and the token leader uses a quadrillion or quadrillions of tokens a month. And I think that will just become the expectation." --- He was at YC Startup School 2026, in conversation Garry Tan From "Y Combinator" YouTube channel, (full video link in comment)

Rohan Paul

20,395 Aufrufe • vor 9 Tagen

Sam Altman on how AI token usage is growing at an exponential rate. "6 and a half years ago, the world token leader was an OpenAI employee using about 100,000 tokens a month. And this seemed ludicrous at the time. The worldwide average per capita was, like, zero. Now, the worldwide average of tokens per month, is like 100,000, and the token leader at OpenAI uses something in the hundreds of billions. If that happens again, which I think it probably will, then in another six and a half years, the average person uses, let’s say, 500 billion tokens a month, and the token leader uses a quadrillion or quadrillions of tokens a month. And I think that will just become the expectation." ---- He was at YC Startup School 2026, in conversation Garry Tan From "Y Combinator" YouTube channel, (full video link in comment)
0:48

Sam Altman on how AI token usage is growing at an exponential rate. "6 and a half years ago, the world token leader was an OpenAI employee using about 100,000 tokens a month. And this seemed ludicrous at the time. The worldwide average per capita was, like, zero. Now, the worldwide average of tokens per month, is like 100,000, and the token leader at OpenAI uses something in the hundreds of billions. If that happens again, which I think it probably will, then in another six and a half years, the average person uses, let’s say, 500 billion tokens a month, and the token leader uses a quadrillion or quadrillions of tokens a month. And I think that will just become the expectation." ---- He was at YC Startup School 2026, in conversation Garry Tan From "Y Combinator" YouTube channel, (full video link in comment)

Rohan Paul

64,715 Aufrufe • vor 1 Monat

Liquid AI (Liquid AI) CEO Ramin Hasani (Ramin) says: "We're bringing the cost of tokens to zero." "The axis was maximizing intelligence at all costs." Foundation models need to optimize across 3 axes: 1.) Intelligence and capability 2.) Efficiency and cost 3.) Substrate: where the intelligence actually runs "Efficiency is not an afterthought. Energy is not abundant." "If you maximize intelligence at all costs, that axis alone is not going to get you to the place that you want to go." "Efficiency and cost of intelligence as a first-class citizen, and not an afterthought." "Where does this intelligence system go? AI is majorly getting hosted in data centers, but you could also bring intelligence on phones, on laptops, on airplanes, on cars." "Imagine if it's not tokens that are actually important. It's just the outcome that actually matters." "Not just thinking about foundation models as the token machines that are generating money and revenue for the foundation model companies that are useless tokens, and getting them into the place where they can actually unlock true value for enterprises."
6:17

Liquid AI (Liquid AI) CEO Ramin Hasani (Ramin) says: "We're bringing the cost of tokens to zero." "The axis was maximizing intelligence at all costs." Foundation models need to optimize across 3 axes: 1.) Intelligence and capability 2.) Efficiency and cost 3.) Substrate: where the intelligence actually runs "Efficiency is not an afterthought. Energy is not abundant." "If you maximize intelligence at all costs, that axis alone is not going to get you to the place that you want to go." "Efficiency and cost of intelligence as a first-class citizen, and not an afterthought." "Where does this intelligence system go? AI is majorly getting hosted in data centers, but you could also bring intelligence on phones, on laptops, on airplanes, on cars." "Imagine if it's not tokens that are actually important. It's just the outcome that actually matters." "Not just thinking about foundation models as the token machines that are generating money and revenue for the foundation model companies that are useless tokens, and getting them into the place where they can actually unlock true value for enterprises."

Molly O’Shea

12,816 Aufrufe • vor 2 Monaten

5 patterns in Text Arena's price–performance Pareto frontier since 2023: 1. GPT-4-level quality is now ~500x lower cost. - From a ~$50 blended price per million tokens in 2023 to ~$0.10 today. 2. The higher-price end is both better and lower-priced since 2023. - The leading Arena score has climbed ~170 points (1,330 → 1,500). While the price of the higher-end frontier models dropped from ~$50 to ~$20 per million tokens. 3. The low-cost end gained the most. - Under $0.20 per million tokens, the best available model went from ~1,000 Arena score in 2023 to ~1,440 today. 4. The low-cost/top performance gap has nearly closed. - In 2023, sub-$0.20 models trailed the leader by ~350 Arena points. Today, ~60. 5. The cast has rotated quite a bit. - - OpenAI set the 2023–24 benchmark. - AI at Meta strengthened the low-cost end in 2024. - Google DeepMind drove the 2025 jump. - Anthropic holds the peak in 2026. - xAI and Chinese labs like DeepSeek AI, Z.ai, Kimi.ai, Xiaomi MiMo, and Qwen are continuing to push the mid-price frontier.
0:26

5 patterns in Text Arena's price–performance Pareto frontier since 2023: 1. GPT-4-level quality is now ~500x lower cost. - From a ~$50 blended price per million tokens in 2023 to ~$0.10 today. 2. The higher-price end is both better and lower-priced since 2023. - The leading Arena score has climbed ~170 points (1,330 → 1,500). While the price of the higher-end frontier models dropped from ~$50 to ~$20 per million tokens. 3. The low-cost end gained the most. - Under $0.20 per million tokens, the best available model went from ~1,000 Arena score in 2023 to ~1,440 today. 4. The low-cost/top performance gap has nearly closed. - In 2023, sub-$0.20 models trailed the leader by ~350 Arena points. Today, ~60. 5. The cast has rotated quite a bit. - - OpenAI set the 2023–24 benchmark. - AI at Meta strengthened the low-cost end in 2024. - Google DeepMind drove the 2025 jump. - Anthropic holds the peak in 2026. - xAI and Chinese labs like DeepSeek AI, Z.ai, Kimi.ai, Xiaomi MiMo, and Qwen are continuing to push the mid-price frontier.

Arena.ai

58,356 Aufrufe • vor 4 Monaten