正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

If your chips and models aren't built in lockstep, you're already behind. At VivaTech, Amazon SVP Peter DeSantis—who leads a new org uniting foundation models and custom chips—explained why they have to develop together, not in sequence. 🤝 "If the chips are not telling the model designers what capabilities... show more

Amazon

5,897,298 subscribers

41,620 次观看 • 2 天前 •via X (Twitter)

科学技术新闻政治

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Chips are the foundation of every AI experience. That's why understanding the hardware behind the software has never mattered more. When you ask a chatbot a question or generate an image, there's a physical chip somewhere doing the heavy lifting. For most of computing history, that was a CPU—the "brain" of a computer—great for general tasks needed to run software and operating systems. AI is more complex: for workloads called training and inference, AI needs to perform trillions of calculations in parallel. That's where AI accelerators come in. Purpose-built accelerators can deliver significantly better performance and efficiency than general-purpose chips. Amazon Web Services Trainium chips are an example—purpose-built for AI training and inference. New chips for a new era. ⬇️

Chips are the foundation of every AI experience. That's why understanding the hardware behind the software has never mattered more. When you ask a chatbot a question or generate an image, there's a physical chip somewhere doing the heavy lifting. For most of computing history, that was a CPU—the "brain" of a computer—great for general tasks needed to run software and operating systems. AI is more complex: for workloads called training and inference, AI needs to perform trillions of calculations in parallel. That's where AI accelerators come in. Purpose-built accelerators can deliver significantly better performance and efficiency than general-purpose chips. Amazon Web Services Trainium chips are an example—purpose-built for AI training and inference. New chips for a new era. ⬇️

Amazon

27,151 次观看 • 1 个月前

We think #GenAI will transform virtually every customer experience. That’s why we’re doubling down on all three layers of what’s called the GenAI “stack.” What do we mean by stack? Simple. Three layers, bottom to top. The “bottom layer” means the infrastructure. Think chips, servers, networking – the software, hardware, and computing power in our data centers used to train, build, and deploy AI models and applications. The “middle layer” is the GenAI sandbox – all the tools, Large Language Models (LLMs), and Foundation Models (FMs) we’ve built into Amazon Bedrock. This lets customers pick the #AI models that best fit their needs to easily and securely customize GenAI applications specific to their business. The “top layer” means the popular agents and assistants that people actually interact with – now including Amazon Q. These are built from the models in the middle layer – in Amazon Bedrock. While other apps might help you write a quick email, or edit an image, Amazon Q’s meant for taking care of business. It’s been engineered to be the most capable GenAI-powered assistant for speeding up software dev and leveraging companies’ internal data. Are you as bullish on the future of GenAI as we are? Listen to our CEO Andy Jassy talk about our investment across the stack.

We think #GenAI will transform virtually every customer experience. That’s why we’re doubling down on all three layers of what’s called the GenAI “stack.” What do we mean by stack? Simple. Three layers, bottom to top. The “bottom layer” means the infrastructure. Think chips, servers, networking – the software, hardware, and computing power in our data centers used to train, build, and deploy AI models and applications. The “middle layer” is the GenAI sandbox – all the tools, Large Language Models (LLMs), and Foundation Models (FMs) we’ve built into Amazon Bedrock. This lets customers pick the #AI models that best fit their needs to easily and securely customize GenAI applications specific to their business. The “top layer” means the popular agents and assistants that people actually interact with – now including Amazon Q. These are built from the models in the middle layer – in Amazon Bedrock. While other apps might help you write a quick email, or edit an image, Amazon Q’s meant for taking care of business. It’s been engineered to be the most capable GenAI-powered assistant for speeding up software dev and leveraging companies’ internal data. Are you as bullish on the future of GenAI as we are? Listen to our CEO Andy Jassy talk about our investment across the stack.

Amazon

47,421 次观看 • 2 年前

AI is not one model. It is a five layer industrial stack: energy, chips and systems, infrastructure software, models, and applications. When America leads across that stack, we lead the next wave of growth.

AI is not one model. It is a five layer industrial stack: energy, chips and systems, infrastructure software, models, and applications. When America leads across that stack, we lead the next wave of growth.

NVIDIA Newsroom

13,522 次观看 • 7 个月前

🚨 Big news for AI innovation: Claude Opus 4 and Claude Sonnet 4, Anthropic's most advanced models, are now available in Amazon Bedrock. These powerful models offer hybrid reasoning, 200K token context windows, and are designed for AI agents. From financial analysis to high-quality writing, to enhanced reasoning, coding, agentic capabilities and more—all with the enterprise-grade security of Amazon Web Services.

🚨 Big news for AI innovation: Claude Opus 4 and Claude Sonnet 4, Anthropic's most advanced models, are now available in Amazon Bedrock. These powerful models offer hybrid reasoning, 200K token context windows, and are designed for AI agents. From financial analysis to high-quality writing, to enhanced reasoning, coding, agentic capabilities and more—all with the enterprise-grade security of Amazon Web Services.

Amazon

123,030 次观看 • 1 年前

Demis Hassabis CEO of DeepMind $GOOG: No one ever has enough chips … and there just aren’t enough compute chips in the world for the demand There’s a reason why $AMZN and $GOOG are running their own semiconductor labs and why we’re building a ridiculous number of semiconductor fabs. AI is projected to have more than 5 billion users by 2030 and the world still not ready for that day Al resources are gonna continue to be undersupplied

Demis Hassabis CEO of DeepMind $GOOG: No one ever has enough chips … and there just aren’t enough compute chips in the world for the demand There’s a reason why $AMZN and $GOOG are running their own semiconductor labs and why we’re building a ridiculous number of semiconductor fabs. AI is projected to have more than 5 billion users by 2030 and the world still not ready for that day Al resources are gonna continue to be undersupplied

Bourbon Capital

163,842 次观看 • 5 个月前

The founder of a $4B inference company says that if you're building agents, foundational models could become your IP. According to Lin Qiao, 90% of the world's data is still private and locked inside applications and enterprises. Foundation models are trained on public internet + labeling company data, which is barely 10% of all the data. And this is why your application and foundation models are misaligned by definition. Companies building agents don't treat models as APIs. They opt for a product-model co-design. - Models continuously learn from your private data - Pick up domain-specific intelligence - Run faster - Cost less - Scale to millions of users

The founder of a $4B inference company says that if you're building agents, foundational models could become your IP. According to Lin Qiao, 90% of the world's data is still private and locked inside applications and enterprises. Foundation models are trained on public internet + labeling company data, which is barely 10% of all the data. And this is why your application and foundation models are misaligned by definition. Companies building agents don't treat models as APIs. They opt for a product-model co-design. - Models continuously learn from your private data - Pick up domain-specific intelligence - Run faster - Cost less - Scale to millions of users

Ivan Burazin

54,964 次观看 • 3 个月前

It currently takes 18-30 months to design and fab a new chip. Jeff Dean sees this shrinking to 3-6 months. And automating design feeds into a recursive loop of improvement: the chips you're designing new chips on can search more and find better designs, which can search more...

It currently takes 18-30 months to design and fab a new chip. Jeff Dean sees this shrinking to 3-6 months. And automating design feeds into a recursive loop of improvement: the chips you're designing new chips on can search more and find better designs, which can search more...

Dwarkesh Patel

108,783 次观看 • 1 年前

Ansem explains the thesis behind Dolphin AI “Dolphin is the provider of the uncensored model that Venice uses” “ChatGPT, Claude, all of the big labs have very strict rules on what you can ask the models, they’re censored very heavily” “We’re not in control of what that censorship is and what they’re telling the models not to say” “One of crypto’s core tenets is not having to rely on some centralized entity deciding what you can and cannot do” “They want the technology to be free and open to everyone. Uncensored models are one way crypto is looking at doing that”

Ansem explains the thesis behind Dolphin AI “Dolphin is the provider of the uncensored model that Venice uses” “ChatGPT, Claude, all of the big labs have very strict rules on what you can ask the models, they’re censored very heavily” “We’re not in control of what that censorship is and what they’re telling the models not to say” “One of crypto’s core tenets is not having to rely on some centralized entity deciding what you can and cannot do” “They want the technology to be free and open to everyone. Uncensored models are one way crypto is looking at doing that”

Market Bubble

20,688 次观看 • 1 个月前

🇺🇸 DAVID SACKS: MODELS ARE LEAPFROGGING EACH OTHER AND BECOMING SPECIALIZED "The model companies are still leapfrogging each other to some degree with their latest versions. And again, that should not be possible if one model achieved rapid takeoff. The other thing you're seeing is that it's not like one model is dominating all the other models. You're seeing that models are developing areas of competitive advantage. They're becoming increasingly specialized." Source: The All-In Podcast David Sacks

🇺🇸 DAVID SACKS: MODELS ARE LEAPFROGGING EACH OTHER AND BECOMING SPECIALIZED "The model companies are still leapfrogging each other to some degree with their latest versions. And again, that should not be possible if one model achieved rapid takeoff. The other thing you're seeing is that it's not like one model is dominating all the other models. You're seeing that models are developing areas of competitive advantage. They're becoming increasingly specialized." Source: The All-In Podcast David Sacks

Mario Nawfal

45,461 次观看 • 10 个月前

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.

I'm running Llama 4 Maverick at 620 t/s! I'm living in the future! Honestly, a large language model running this fast is something straight out of a sci-fi movie. Speeds like this will enable a whole new world of applications that aren't possible today. For reference, GPT-4o, which is probably the most popular OpenAI model, runs between 60 and 110 t/s. The secret here: I'm not running AI at Meta's Llama 4 Maverick on a GPU. I'm using the SambaNova Cloud (my sponsor) and their custom SN40L chips. They are optimized from the ground up for running AI workflows. Right now, SambaNova Cloud runs DeepSeek, Qwen, Whisper, and the entire family of Llama models on these chips. You can check the speed of each of these models using SambaNova Cloud's Playground (see the attached video). It's completely free, and that's how I'm measuring their speeds. For example, I also tried DeepSeek R1 (the latest version from May) and, oh boy! DeepSeek R1 is a huge 671B parameter model. It's probably the best open reasoning model in the world, and it runs at 140 tokens per second! !!! Inference time on an SN40L is night and day from what you'll get from a GPU. Here is why this is big: If you are running an agentic workflow that uses multiple models simultaneously on a GPU, it will need to swap models in and out of memory (because not every model fits). A single SNL40 chip can simultaneously hold over 100 models (trillions of parameters) in memory. If you are using open models, try the SambaCloud API to see what lightning speed looks like. Here is how: 1. Create a free account at: 2. Check the QuickStart guide: If you try the playground, check the speed you're getting with Llama 4 and DeepSeek, and post the results below. I've seen much higher numbers than I posted here, so I'm curious to see whether geography affects the speed.

Santiago

34,148 次观看 • 1 年前

Wood chips being found in Dave's Killer Bread Not small wood shavings, not little pieces. These are pretty large wood chips (shown) and they are throughout the entire loaf This has already been reported to the FDA Dave's Killer Bread is the top-selling organic sliced bread brand in the US

Wall Street Apes

1,499,491 次观看 • 2 个月前

They are not there, but if you have a beefy mac, Qwen3-Coder-Next has some applications due to its speed and decent tool calling capabilities. This video is in real time. (Also plenty of local models for STT etc.). They are not to be completely dismissed!

They are not there, but if you have a beefy mac, Qwen3-Coder-Next has some applications due to its speed and decent tool calling capabilities. This video is in real time. (Also plenty of local models for STT etc.). They are not to be completely dismissed!

Armin Ronacher ⇌

36,110 次观看 • 3 个月前

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:

This is the fastest I've seen Llama 3.3 running anywhere! Llama 3.3 70B running at 652 t/s is lightning fast. And if you want Llama 3.1, here are the speeds I was able to get: • Llama 3.1 8B: 1006 t/s • Llama 3.1 70B: 709 t/s • Llama 3.1 405B: 206 t/s (You can access all of these models for free! See the link below.) This speed is incredible, but the interesting part is what's happening behind the scenes: These models aren't running on a GPU! In this video, I'm using the SambaNova cloud to access these models. They built a custom chip (SN40L) optimized for AI workflows. A single SN40L chip can hold hundreds of models (trillions of parameters) in memory! The speed alone is a huge deal, but the big advantage is for agentic workflows running multiple specialized models. A GPU can only host a single model and switch (unload and load) to a different model if necessary. An SN40L, on the other hand, can host every model at once, making it much faster. Here is the video where you can see how fast these chips are:

Santiago

97,505 次观看 • 1 年前

BANNON: This is the sitting president accusing his predecessor of treasonous conspiracy. Never happened before. It has to be adjudicated with full transparency. If they are not guilty, say so and move on. But if they are, and the evidence is overwhelming, then let the chips fall where they may.

BANNON: This is the sitting president accusing his predecessor of treasonous conspiracy. Never happened before. It has to be adjudicated with full transparency. If they are not guilty, say so and move on. But if they are, and the evidence is overwhelming, then let the chips fall where they may.

Grace Chong, MBI

61,939 次观看 • 11 个月前

Dojo 2, Dojo 3 Chips and AI5 Chips Are Critical for Tesla and Battle for Future of AI. Tesla and XAI are buying about one million AI GPUs this year. This would be about 20% of the entire Nvidia production of B200/B300 chips that will be made in 2025. If XAI were to buy a mix of half Dojo 2/3 chips in 2025 and 2026 this could make them bigger than AMD. AMD is making about 400,000 AI GPU chips. Tesla will switch to AI5 chips for driving its cars but those chips will also act as inference chips with about 500 Teraflops of compute. Tesla could make 3-4 million AI5 chips in 2026. This would make Tesla the largest maker and user of AI inference chips in the world. Tesla could become the number 2 maker of AI training chips just by supplying a lot of those chips to XAI and Tesla internal use. Tesla is and will be the number one maker and user of AI inference chips. Sawyer Merritt Elon Musk Randy Kirk Herbert Ong Warren Redlich - Chasing Dreams 🇺🇸 Ale𝕏andra Merz 🇺🇲 Ray Emmet Peppers Bradford Ferguson

Dojo 2, Dojo 3 Chips and AI5 Chips Are Critical for Tesla and Battle for Future of AI. Tesla and XAI are buying about one million AI GPUs this year. This would be about 20% of the entire Nvidia production of B200/B300 chips that will be made in 2025. If XAI were to buy a mix of half Dojo 2/3 chips in 2025 and 2026 this could make them bigger than AMD. AMD is making about 400,000 AI GPU chips. Tesla will switch to AI5 chips for driving its cars but those chips will also act as inference chips with about 500 Teraflops of compute. Tesla could make 3-4 million AI5 chips in 2026. This would make Tesla the largest maker and user of AI inference chips in the world. Tesla could become the number 2 maker of AI training chips just by supplying a lot of those chips to XAI and Tesla internal use. Tesla is and will be the number one maker and user of AI inference chips. Sawyer Merritt Elon Musk Randy Kirk Herbert Ong Warren Redlich - Chasing Dreams 🇺🇸 Ale𝕏andra Merz 🇺🇲 Ray Emmet Peppers Bradford Ferguson

nextbigfuture

23,809 次观看 • 1 年前

#WATCH | Semicon India 2025 | Sandeep Kumar, CEO, L&T Semiconductor Technologies, says, "We are building a new security protocol chip. All the chips currently in use in passports, Aadhaar cards, and payment cards are imported. There are high chances of security and data breaches. At L&T, we will develop those chips for the security protocol. These chips will be released within 1.5 years. These chips will be used in passports, credit cards, and Aadhar cards, which will become electronic, and all payment gateways... The government already has plans to make new passports with electronic chips, so as soon as the chips are ready, passports will be made using these chips..."

#WATCH | Semicon India 2025 | Sandeep Kumar, CEO, L&T Semiconductor Technologies, says, "We are building a new security protocol chip. All the chips currently in use in passports, Aadhaar cards, and payment cards are imported. There are high chances of security and data breaches. At L&T, we will develop those chips for the security protocol. These chips will be released within 1.5 years. These chips will be used in passports, credit cards, and Aadhar cards, which will become electronic, and all payment gateways... The government already has plans to make new passports with electronic chips, so as soon as the chips are ready, passports will be made using these chips..."

ANI

54,188 次观看 • 10 个月前

At RAISE Summit in Paris, one theme stood out: every layer of the AI stack is up for grabs. Startups are challenging incumbents in chips, cloud infrastructure, data centers, and foundation models alike.

At RAISE Summit in Paris, one theme stood out: every layer of the AI stack is up for grabs. Startups are challenging incumbents in chips, cloud infrastructure, data centers, and foundation models alike.

TIME

25,401 次观看 • 2 天前

Nvidia CEO Jensen Huang: There’s a belief that somehow if we provided American chips to foreign countries — and specifically China — that the Chinese military would be building their military, building airplanes and aircraft carriers using American chips. First of all, they have plenty of their own chips, and their companies are formidable. The idea that China does not have a technology industry — we now know — is lunacy. Source: TIME

Nvidia CEO Jensen Huang: There’s a belief that somehow if we provided American chips to foreign countries — and specifically China — that the Chinese military would be building their military, building airplanes and aircraft carriers using American chips. First of all, they have plenty of their own chips, and their companies are formidable. The idea that China does not have a technology industry — we now know — is lunacy. Source: TIME

Clash Report

62,118 次观看 • 6 个月前

“These models and applications are very power hungry, but there are things we can do together.” – Amazon CEO Andy Jassy on offsetting the energy impacts of Gen AI to remain power performant. Listen to find out 👇 : Why using Amazon Web Services is more energy efficient than on-premises Why Tranium helps (by nearly 30%!) Why Inferentia is so much more energy efficient What we’re doing about our 2030 water promise

“These models and applications are very power hungry, but there are things we can do together.” – Amazon CEO Andy Jassy on offsetting the energy impacts of Gen AI to remain power performant. Listen to find out 👇 : Why using Amazon Web Services is more energy efficient than on-premises Why Tranium helps (by nearly 30%!) Why Inferentia is so much more energy efficient What we’re doing about our 2030 water promise

Amazon

46,582 次观看 • 1 年前

Chips are being embedded into your clothing labels The chips are mainly RFID, Radio Frequency Identification, chips being embedded into the labels If you look closely or hold the labels up to light you can clearly see the chip, not all labels have them but many retailers are using them RFID chips are used in many huge retailers in America for - Inventory tracking - Self-checkout, you place items in a bin and the system reads all tags automatically - Anti-theft and security They contain a microchip and antenna but are not “tracking microchips” however I’d still recommend removing them They’ve been out for years but are being used by more and more major retailers so people are staying to notice

Chips are being embedded into your clothing labels The chips are mainly RFID, Radio Frequency Identification, chips being embedded into the labels If you look closely or hold the labels up to light you can clearly see the chip, not all labels have them but many retailers are using them RFID chips are used in many huge retailers in America for - Inventory tracking - Self-checkout, you place items in a bin and the system reads all tags automatically - Anti-theft and security They contain a microchip and antenna but are not “tracking microchips” however I’d still recommend removing them They’ve been out for years but are being used by more and more major retailers so people are staying to notice

Wall Street Apes

175,582 次观看 • 2 天前