Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Chinese scientists unveil a superfast analog chip, 1000x faster than top digital processors. It solves complex maths for AI & science, outperforming even NVIDIA's GPUs in tests. The race for the best computing power is entering its next phase.

Chubby♨️

116,899 subscribers

66,056 views • 7 months ago •via X (Twitter)

Science & Technology

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Groq's LPU is faster than GPUs, handling requests and responding more quickly. Groq's LPUs don't need speedy data delivery like Nvidia GPUs do because they don't have HBM in their system. They use SRAM, which is about 20 times faster than what GPUs use. Since inference runs use way less data than model training, Groq's LPU is more energy-efficient. It reads less from external memory and uses less power than a Nvidia GPU for inference tasks. The LPU works differently from GPUs. It uses a Temporal Instruction Set Computer architecture, so it doesn't have to reload data from memory as often as GPUs do with High Bandwidth Memory (HBM). This helps avoid issues with HBM shortages and keeps costs down. If Groq's LPU is used in places that do AI processing, you might not need special storage for Nvidia GPUs. The LPU doesn't demand super-fast storage like GPUs do. Groq is claiming that its technology could replace GPUs in AI tasks with its powerful chip and software.

Groq's LPU is faster than GPUs, handling requests and responding more quickly. Groq's LPUs don't need speedy data delivery like Nvidia GPUs do because they don't have HBM in their system. They use SRAM, which is about 20 times faster than what GPUs use. Since inference runs use way less data than model training, Groq's LPU is more energy-efficient. It reads less from external memory and uses less power than a Nvidia GPU for inference tasks. The LPU works differently from GPUs. It uses a Temporal Instruction Set Computer architecture, so it doesn't have to reload data from memory as often as GPUs do with High Bandwidth Memory (HBM). This helps avoid issues with HBM shortages and keeps costs down. If Groq's LPU is used in places that do AI processing, you might not need special storage for Nvidia GPUs. The LPU doesn't demand super-fast storage like GPUs do. Groq is claiming that its technology could replace GPUs in AI tasks with its powerful chip and software.

Brian Roemmele

87,188 views • 2 years ago

NVIDIA became the default choice for some of the most ambitious AI startups — like Mistral AI. How? Jensen Huang shares the philosophy that made it possible: “We’re not a chip company. We’re a developer-first computing platform.” Hear how that mindset shaped NVIDIA's role in the AI ecosystem. 👇

NVIDIA became the default choice for some of the most ambitious AI startups — like Mistral AI. How? Jensen Huang shares the philosophy that made it possible: “We’re not a chip company. We’re a developer-first computing platform.” Hear how that mindset shaped NVIDIA's role in the AI ecosystem. 👇

a16z

76,941 views • 1 year ago

"Just like maths...was the perfect description language for physics, I think that AI is potentially the perfect description language for biology." Demis Hassabis envisions a new era of "digital biology" where AI helps us understand life's complex information processing. His dream? "Virtual cells" for faster, more efficient scientific discovery.

"Just like maths...was the perfect description language for physics, I think that AI is potentially the perfect description language for biology." Demis Hassabis envisions a new era of "digital biology" where AI helps us understand life's complex information processing. His dream? "Virtual cells" for faster, more efficient scientific discovery.

vitrupo

28,827 views • 1 year ago

“In the next 10 years we are going to be entering what I feel like is a new renaissance.” Nobel Prize laureate Demis Hassabis was awarded the Nobel Prize for using AI to predict the structure of proteins. For him, AI is a tool that will help scientists make even more discoveries in the years to come. Hassabis took part in our Nobel Prize Dialogue ‘The Future of Science With AI’ which discussed how AI might transform science in the future. Watch the full event at

“In the next 10 years we are going to be entering what I feel like is a new renaissance.” Nobel Prize laureate Demis Hassabis was awarded the Nobel Prize for using AI to predict the structure of proteins. For him, AI is a tool that will help scientists make even more discoveries in the years to come. Hassabis took part in our Nobel Prize Dialogue ‘The Future of Science With AI’ which discussed how AI might transform science in the future. Watch the full event at

The Nobel Prize

156,797 views • 2 months ago

From Nvidia's new AI chip for China to Apple's push for blood oxygen monitoring on many more of its watches, here is a roundup of the week's most trending stories in tech

From Nvidia's new AI chip for China to Apple's push for blood oxygen monitoring on many more of its watches, here is a roundup of the week's most trending stories in tech

Reuters

40,620 views • 11 months ago

How sophisticated is the maths in AI? Go on, have a guess. Because guessing is what AI itself would do. Terry Tao introduces his Oxford Mathematics London Lecture, a journey through the potential for AI in science & maths. Watch from 5pm BST, Wed 7 August:

How sophisticated is the maths in AI? Go on, have a guess. Because guessing is what AI itself would do. Terry Tao introduces his Oxford Mathematics London Lecture, a journey through the potential for AI in science & maths. Watch from 5pm BST, Wed 7 August:

Oxford Mathematics

62,121 views • 2 years ago

AI Speaker is shipping! The world’s first AI smart speaker designed for #DePIN is now on its way to your doorstep. Powered by the “Boom Chip” for on-device AI and DePIN edge computing, this Web3 speaker is more than just smart — it’s a gateway to the decentralized future.

AI Speaker is shipping! The world’s first AI smart speaker designed for #DePIN is now on its way to your doorstep. Powered by the “Boom Chip” for on-device AI and DePIN edge computing, this Web3 speaker is more than just smart — it’s a gateway to the decentralized future.

MelosBoom

31,448 views • 1 year ago

Nvidia received Beijing's approval to sell its H200 AI chips in China and is preparing a Groq AI chip version for the Chinese market, sources said

Nvidia received Beijing's approval to sell its H200 AI chips in China and is preparing a Groq AI chip version for the Chinese market, sources said

Reuters Business

14,904 views • 4 months ago

Microsoft just had the biggest breakthrough in Quantum computing! Meet Majorana 1 - a chip that can perform incredibly complex calculations that would be impossible for even the largest classical computers… and I got to see it in person!!

Microsoft just had the biggest breakthrough in Quantum computing! Meet Majorana 1 - a chip that can perform incredibly complex calculations that would be impossible for even the largest classical computers… and I got to see it in person!!

iJustine

569,078 views • 1 year ago

🇸🇦 SAUDI ARABIA GOES SUPERCOMPUTER: NVIDIA DEAL TO BUILD AI MEGASCALE Saudi Arabia partnership with Nvidia to build one of the world’s largest AI infrastructures is a monumental step in the proliferation of Artificial Intelligence. The Kingdom will deploy 18,000 Nvidia GB300 superchips in Phase 1, aiming for 500 megawatts of compute power within five years. This move will undoubtedly position Saudi Arabia as a regional AI hub amid a global race for computing supremacy. It will support sovereign AI models and digital twin simulations across industries. As Vision 2030 shifts the Kingdom beyond oil, AI is now its new strategic resource—and it's going all in. Source: AI Magazine

🇸🇦 SAUDI ARABIA GOES SUPERCOMPUTER: NVIDIA DEAL TO BUILD AI MEGASCALE Saudi Arabia partnership with Nvidia to build one of the world’s largest AI infrastructures is a monumental step in the proliferation of Artificial Intelligence. The Kingdom will deploy 18,000 Nvidia GB300 superchips in Phase 1, aiming for 500 megawatts of compute power within five years. This move will undoubtedly position Saudi Arabia as a regional AI hub amid a global race for computing supremacy. It will support sovereign AI models and digital twin simulations across industries. As Vision 2030 shifts the Kingdom beyond oil, AI is now its new strategic resource—and it's going all in. Source: AI Magazine

Mario Nawfal

188,830 views • 1 year ago

The GPU market is more complex than it seems. 🧐 Keith Townsend (Keith Townsend) from The Futurum Group, provides a nuanced perspective on H100/H200 demand from Google Cloud Next, highlighting the ongoing importance of these chips in the AI infrastructure landscape. #AI #GPUs

The GPU market is more complex than it seems. 🧐 Keith Townsend (Keith Townsend) from The Futurum Group, provides a nuanced perspective on H100/H200 demand from Google Cloud Next, highlighting the ongoing importance of these chips in the AI infrastructure landscape. #AI #GPUs

Six Five Media

31,132 views • 1 year ago

I sat down with NVIDIA Sam Stanwyck (Sam Stanwyck) at ISC to talk about one of the most misunderstood areas in computing: quantum. We discussed why NVIDIA is not building a quantum computer, but instead the infrastructure around it: GPUs, AI, CUDA-Q, quantum simulation, control systems, error correction, and the path toward hybrid CPU/GPU/QPU supercomputing. The key takeaway: Quantum will likely not arrive as a standalone magic machine. It will arrive as part of a much larger accelerated computing stack, combining classical supercomputers, AI, GPUs, and quantum processors to tackle problems in chemistry, materials science, energy, optimization, and fundamental science. Full episode with Sam Stanwyck is now live.

I sat down with NVIDIA Sam Stanwyck (Sam Stanwyck) at ISC to talk about one of the most misunderstood areas in computing: quantum. We discussed why NVIDIA is not building a quantum computer, but instead the infrastructure around it: GPUs, AI, CUDA-Q, quantum simulation, control systems, error correction, and the path toward hybrid CPU/GPU/QPU supercomputing. The key takeaway: Quantum will likely not arrive as a standalone magic machine. It will arrive as part of a much larger accelerated computing stack, combining classical supercomputers, AI, GPUs, and quantum processors to tackle problems in chemistry, materials science, energy, optimization, and fundamental science. Full episode with Sam Stanwyck is now live.

Chubby♨️

22,673 views • 1 month ago

📁 Neil deGrasse Tyson, astrophysicist and science communicator, says quantum computing completely breaks our intuition. With 53 qubits, you already run 10¹⁶ states in parallel, more combinations than the mind can grasp. Scale it further and you surpass the number of atoms in the universe. This is not faster computing, it is a different kind of power. When parallelism explodes, the idea of limits collapses.

📁 Neil deGrasse Tyson, astrophysicist and science communicator, says quantum computing completely breaks our intuition. With 53 qubits, you already run 10¹⁶ states in parallel, more combinations than the mind can grasp. Scale it further and you surpass the number of atoms in the universe. This is not faster computing, it is a different kind of power. When parallelism explodes, the idea of limits collapses.

Jon Hernandez

38,758 views • 6 months ago

Deep in the mountains of Seedworld An ancient power awaits. For each 120.000 $SWORLD you stake You'll be entering a race. In it, you will find great value As well as the power to shape. Await the next sign. 🌋

Deep in the mountains of Seedworld An ancient power awaits. For each 120.000 $SWORLD you stake You'll be entering a race. In it, you will find great value As well as the power to shape. Await the next sign. 🌋

Seedworld

309,746 views • 1 year ago

Elon Musk reveals the real limiting factor for AI is not chips anymore, it’s electricity AI chip production is growing exponentially, but global electricity supply is only increasing 3–4% per year. We’re reaching the point where we produce more chips than we can actually turn on Power is the ultimate bottleneck in 2026. Data centers now face up to 7-year waits for grid connections, leaving many GPUs sitting idle Energy is the new oil of the AI race

Elon Musk reveals the real limiting factor for AI is not chips anymore, it’s electricity AI chip production is growing exponentially, but global electricity supply is only increasing 3–4% per year. We’re reaching the point where we produce more chips than we can actually turn on Power is the ultimate bottleneck in 2026. Data centers now face up to 7-year waits for grid connections, leaving many GPUs sitting idle Energy is the new oil of the AI race

X Freeze

33,904 views • 4 months ago

.Glinert 🇺🇸 🏭 (Co-Founder & CEO of Sphere Semi) on how they're using AI to change how chips are designed: “CPUs and GPUs get all the attention, but analog is a third of the trillion-dollar chip industry and it’s critical for communications, warfare, and more.” “We think human chip design is coming to an end. In analog, it’s still done by hand and we’re putting that to an end.”

.Glinert 🇺🇸 🏭 (Co-Founder & CEO of Sphere Semi) on how they're using AI to change how chips are designed: “CPUs and GPUs get all the attention, but analog is a third of the trillion-dollar chip industry and it’s critical for communications, warfare, and more.” “We think human chip design is coming to an end. In analog, it’s still done by hand and we’re putting that to an end.”

TBPN

58,744 views • 10 months ago

💨 AI shouldn’t be slowed down by its own infrastructure. Introducing Cottonia — a smarter, lighter cloud acceleration layer for AI. We use distributed computing and intelligent scheduling to make AI execution faster & cheaper. ♻️The future of intelligent compute is here.

💨 AI shouldn’t be slowed down by its own infrastructure. Introducing Cottonia — a smarter, lighter cloud acceleration layer for AI. We use distributed computing and intelligent scheduling to make AI execution faster & cheaper. ♻️The future of intelligent compute is here.

Cottonia

13,668 views • 6 months ago

The next computing shift is already underway — and it’s happening faster than even the people building it expected. Meta CTO Boz joins David George to talk about how the team is rooted in solving real-world problems — and how this new wave of AI is unlike any before, ushering us towards a future beyond the smartphone. Boz says AI isn’t just making things faster or cheaper, it’s making everything better. From interfaces to interaction design, AI is transforming how we build — and how we compute. For more on the future of computing, wearables, and AI-native UX, watch the full conversation with Boz and David:

The next computing shift is already underway — and it’s happening faster than even the people building it expected. Meta CTO Boz joins David George to talk about how the team is rooted in solving real-world problems — and how this new wave of AI is unlike any before, ushering us towards a future beyond the smartphone. Boz says AI isn’t just making things faster or cheaper, it’s making everything better. From interfaces to interaction design, AI is transforming how we build — and how we compute. For more on the future of computing, wearables, and AI-native UX, watch the full conversation with Boz and David:

a16z

17,480 views • 1 year ago

ELON: EVEN OPEN-SOURCE AI IS GETTING BETTER HUMAN TESTS THAN HUMANS "We're getting to the point where even open-source AI can pass human captcha tests: 'Are you a human? Identify all the traffic lights in this picture.' You're like, okay, it's going to have no problem doing that. In fact, it'll do it better than a human and faster than a human. So, at the point at which it's better passing human tests than humans, then, what tests actually make sense?" Source: Interview with Rishi Sunak, November 2023

ELON: EVEN OPEN-SOURCE AI IS GETTING BETTER HUMAN TESTS THAN HUMANS "We're getting to the point where even open-source AI can pass human captcha tests: 'Are you a human? Identify all the traffic lights in this picture.' You're like, okay, it's going to have no problem doing that. In fact, it'll do it better than a human and faster than a human. So, at the point at which it's better passing human tests than humans, then, what tests actually make sense?" Source: Interview with Rishi Sunak, November 2023

Mario Nawfal

42,071 views • 1 year ago