正在加载视频...

视频加载失败

返回首页

Introducing SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Pre-Training Corpora What lies within a trillion-scale pre-training corpus? Can you truly guarantee your benchmarks are uncontaminated simply because there are no exact string matches? Alongside several research institutions in Japan, Sakana AI is proud to have collaborated...

SakanaAILabs's profile picture

Sakana AI

72,726 subscribers

88,794 次观看 • 4 个月前 •via X (Twitter)

教育科学技术
LIVE

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

Still following your human intuition to mix corpora from different sources for language model pre-training 🧠? Everyone says that data mixture has a big impact on model performance, but how - and why🕵️? Did you know that web corpora are actually highly impactful for downstream tasks 🏆? Let's check out our preprint "RegMix: Data Mixture as Regression for Language Model Pre-training" 📄 🔬In this paper, we've proposed an automatic data mixture method RegMix that achieves a 6.3% improvement over human selection on the widely used HellaSwag benchmark - and it only needs a 2% extra training FLOPs! 📈 Details in the thread 🧵
0:34

Still following your human intuition to mix corpora from different sources for language model pre-training 🧠? Everyone says that data mixture has a big impact on model performance, but how - and why🕵️? Did you know that web corpora are actually highly impactful for downstream tasks 🏆? Let's check out our preprint "RegMix: Data Mixture as Regression for Language Model Pre-training" 📄 🔬In this paper, we've proposed an automatic data mixture method RegMix that achieves a 6.3% improvement over human selection on the widely used HellaSwag benchmark - and it only needs a 2% extra training FLOPs! 📈 Details in the thread 🧵

Qian Liu

54,961 次观看 • 1 年前

To prove how scalable Upstash Vector is, we indexed the entire Wikipedia in 11 languages (144m vectors) in a single DB. ◆ Over 700GB of data ◆ Fast semantic search ◆ Chat with Wikipedia We got your back for a fast app that scales🫡 Quick demo 👇
1:32

To prove how scalable Upstash Vector is, we indexed the entire Wikipedia in 11 languages (144m vectors) in a single DB. ◆ Over 700GB of data ◆ Fast semantic search ◆ Chat with Wikipedia We got your back for a fast app that scales🫡 Quick demo 👇

Upstash

31,069 次观看 • 1 年前

The infrastructure for working with video data at scale barely exists. Grass Video Search is our first step in changing that. You can now find anything in a video based on the contents of its frames, not just transcripts or tags. Over the last several months, the Grass Foundation has quietly assembled one of the largest video datasets in history: over a billion videos from across the public web, and is segmenting them by scene with VLM annotations of keyframes. This dataset is now searchable, so you can find the exact moment a visual, pattern, or concept appears. Curation has always been the bottleneck in video analysis. We built a system that automates it. This is just the beginning. Over 4 billion additional videos are queued for processing, and the index is growing daily. We’re opening access to a limited set of collaborators. If you’re a researcher, journalist, or nonprofit working with video, archives, or large-scale analysis, we’d love to hear from you.
0:45

The infrastructure for working with video data at scale barely exists. Grass Video Search is our first step in changing that. You can now find anything in a video based on the contents of its frames, not just transcripts or tags. Over the last several months, the Grass Foundation has quietly assembled one of the largest video datasets in history: over a billion videos from across the public web, and is segmenting them by scene with VLM annotations of keyframes. This dataset is now searchable, so you can find the exact moment a visual, pattern, or concept appears. Curation has always been the bottleneck in video analysis. We built a system that automates it. This is just the beginning. Over 4 billion additional videos are queued for processing, and the index is growing daily. We’re opening access to a limited set of collaborators. If you’re a researcher, journalist, or nonprofit working with video, archives, or large-scale analysis, we’d love to hear from you.

Grass

469,986 次观看 • 1 年前

NEW: Elon Musk says he thinks he can reduce the deficit by $1 Trillion next year, says it will cause a cascading effect for Americans. “I think we can take a trillion dollars out of the out of the deficit next year.” “We have a $2 trillion deficit, which is a far in excess of economic growth. If we can get that deficit in half, from 2 trillion to 1 trillion, and we can get the economic growth to match that 1 trillion growth in the money supply, that means there will be no inflation.” “Interest rates will drop, credit card interest down, mortgage interest down, car payment interest down, and prices at the store stay the same.” “That is a great outcome for people and I think we can do it.”
2:07

NEW: Elon Musk says he thinks he can reduce the deficit by $1 Trillion next year, says it will cause a cascading effect for Americans. “I think we can take a trillion dollars out of the out of the deficit next year.” “We have a $2 trillion deficit, which is a far in excess of economic growth. If we can get that deficit in half, from 2 trillion to 1 trillion, and we can get the economic growth to match that 1 trillion growth in the money supply, that means there will be no inflation.” “Interest rates will drop, credit card interest down, mortgage interest down, car payment interest down, and prices at the store stay the same.” “That is a great outcome for people and I think we can do it.”

Collin Rugg

4,342,838 次观看 • 1 年前

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin
0:40

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin

机器之心 JIQIZHIXIN

18,126 次观看 • 7 个月前

At Omnipresent, We incubate. We build. Introducing our first product a marketing tool that enables an entirely different way of operating. Our CTO taro developed an in-house stack using OpenClaw🦞, Kimi.ai, and Claude built to move faster, hit harder, and scale beyond what off-the-shelf tools allow. What it unlocks: – Sharper, faster marketing execution – Distribution that scales beyond the obvious channels – High-leverage creative angles that compound - Marketing analytics and data you can trust And this is just an experiment. We are working on AI dashboard with new layers of insights , network scale effects , and a UI/UX bar that keeps moving up. This is what building from the inside out looks like.
0:29

At Omnipresent, We incubate. We build. Introducing our first product a marketing tool that enables an entirely different way of operating. Our CTO taro developed an in-house stack using OpenClaw🦞, Kimi.ai, and Claude built to move faster, hit harder, and scale beyond what off-the-shelf tools allow. What it unlocks: – Sharper, faster marketing execution – Distribution that scales beyond the obvious channels – High-leverage creative angles that compound - Marketing analytics and data you can trust And this is just an experiment. We are working on AI dashboard with new layers of insights , network scale effects , and a UI/UX bar that keeps moving up. This is what building from the inside out looks like.

Omnipresent

14,684 次观看 • 2 个月前

I cant believe this guy just made a permanent solution to context bloat and open sourced it all! when we tested this tool (Context+) for solving an issue on the OpenCode repository, the agent using this tool used ~6.5k fewer tokens, found the code and fixed it in half the time! the results were surprising: 6 to 10k tokens saved per prompt, completed task in ~2 minutes while the agent running without the tool took ~4 mins for the same and got stuck in loops bro built an entire beast by using all the modern tools that we could think of: undo trees, semantic search by meaning (by haskellforall), advanced refactoring, blast radius, advanced file context trees, restore points... i can keep going on semantic code search and context trees are the future of agentic coding and this tool proves it the feature i loved the most is semantic search and how it gets things done 2x faster with least possible tokens it makes an agent that actually knows what it’s doing and not just guessing, it makes meaning from your code similar to RAG. if you aren't optimizing your context, you are just burning money the developer says this tool is still under development, it can have unexpected behavior and the docs need updates but the video shows the reality of how fast it can be github: get here:
0:46

I cant believe this guy just made a permanent solution to context bloat and open sourced it all! when we tested this tool (Context+) for solving an issue on the OpenCode repository, the agent using this tool used ~6.5k fewer tokens, found the code and fixed it in half the time! the results were surprising: 6 to 10k tokens saved per prompt, completed task in ~2 minutes while the agent running without the tool took ~4 mins for the same and got stuck in loops bro built an entire beast by using all the modern tools that we could think of: undo trees, semantic search by meaning (by haskellforall), advanced refactoring, blast radius, advanced file context trees, restore points... i can keep going on semantic code search and context trees are the future of agentic coding and this tool proves it the feature i loved the most is semantic search and how it gets things done 2x faster with least possible tokens it makes an agent that actually knows what it’s doing and not just guessing, it makes meaning from your code similar to RAG. if you aren't optimizing your context, you are just burning money the developer says this tool is still under development, it can have unexpected behavior and the docs need updates but the video shows the reality of how fast it can be github: get here:

forloop

225,774 次观看 • 3 个月前

The race to the multi-trillion-dollar tokenized economy has begun. And Nexera is providing the infrastructure behind it. Our modular architecture and white-label suite, powered by Evergon , make launching a tokenized market as easy as opening a store on Shopify. No code. No friction. No barriers. From institutions to startups, anyone can go to market in minutes, fully compliant. Over $1B in pipeline already… and growing fast. Build fast. Launch compliant. Scale globally. This is Nexera. $NXRA #Tokenization #Nexera #Evergon #RWA #Infrastructure #DigitalAssets #Web3
2:35

The race to the multi-trillion-dollar tokenized economy has begun. And Nexera is providing the infrastructure behind it. Our modular architecture and white-label suite, powered by Evergon , make launching a tokenized market as easy as opening a store on Shopify. No code. No friction. No barriers. From institutions to startups, anyone can go to market in minutes, fully compliant. Over $1B in pipeline already… and growing fast. Build fast. Launch compliant. Scale globally. This is Nexera. $NXRA #Tokenization #Nexera #Evergon #RWA #Infrastructure #DigitalAssets #Web3

Nexera

12,784 次观看 • 8 个月前

SpiRobs morphologically replicate the logarithmic spiral that is ubiquitous in natural organisms. They are easy and fast to build across scales via 3D printing. They are actuated by cables, which allow for fast and life-like movements. Besides, a single robot can handle a wide variety of objects (in terms of size, shape, and weight). A key to this is a bioinspired grasping strategy that is found in the octopus. Finally, this work demonstrates a wide range of prototypes, including a miniaturized gripper, a manipulator mounted on a drone, and multi-robot arrays that can grasp in a tendril-like fashion. Authors: Zhanchi Wang, Nikolaos M. Freris, and Xi Wei University of Science and Technology of China Video Credit: ———————————————— Stay ahead of the curve! Follow more cutting edge robotics content, including research, reports and articles on Wevolver:
0:33

SpiRobs morphologically replicate the logarithmic spiral that is ubiquitous in natural organisms. They are easy and fast to build across scales via 3D printing. They are actuated by cables, which allow for fast and life-like movements. Besides, a single robot can handle a wide variety of objects (in terms of size, shape, and weight). A key to this is a bioinspired grasping strategy that is found in the octopus. Finally, this work demonstrates a wide range of prototypes, including a miniaturized gripper, a manipulator mounted on a drone, and multi-robot arrays that can grasp in a tendril-like fashion. Authors: Zhanchi Wang, Nikolaos M. Freris, and Xi Wei University of Science and Technology of China Video Credit: ———————————————— Stay ahead of the curve! Follow more cutting edge robotics content, including research, reports and articles on Wevolver:

Wevolver

23,154 次观看 • 1 年前

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation paper page: The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.
4:47

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation paper page: The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.

AK

18,871 次观看 • 2 年前

Today morning my supplementary question no 1 on Young India Skills University. I'm proud to say that under the able and dynamic leadership of our Hon'ble Chief Minister Revanth Reddy, he is ambitiously setting up the Young India Skills University in Hyderabad, with the objective of providing job opportunities to the students and imparting skill development training according to the requirements of the local industries that are in demand in collaboration with various companies in the PPP model. A total of 17 priority sectors have been identified, and we will be providing training to 2 thousand people in the first year of 2025-26. We will train 20,000 people every year, and the state has allocated Rs. 100 crore on its part for the same and further wants to create a corpus fund of about Rs. 300-500 crores. Now my question to the ministry is if there is any proposal from the central government to adequately fund this Skills University. This will make it a grand success and will lead to the realisation of the goal of a developed nation by 2047. #YoungIndiaSkillUniversity #Telangana #TelanganaRising
1:12

Today morning my supplementary question no 1 on Young India Skills University. I'm proud to say that under the able and dynamic leadership of our Hon'ble Chief Minister Revanth Reddy, he is ambitiously setting up the Young India Skills University in Hyderabad, with the objective of providing job opportunities to the students and imparting skill development training according to the requirements of the local industries that are in demand in collaboration with various companies in the PPP model. A total of 17 priority sectors have been identified, and we will be providing training to 2 thousand people in the first year of 2025-26. We will train 20,000 people every year, and the state has allocated Rs. 100 crore on its part for the same and further wants to create a corpus fund of about Rs. 300-500 crores. Now my question to the ministry is if there is any proposal from the central government to adequately fund this Skills University. This will make it a grand success and will lead to the realisation of the goal of a developed nation by 2047. #YoungIndiaSkillUniversity #Telangana #TelanganaRising

Kiran Kumar Chamala

71,229 次观看 • 1 年前

Progressively Optimized Local Radiance Fields for Robust View Synthesis paper page: present an algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video. The task poses two core challenges. First, most existing radiance field reconstruction approaches rely on accurate pre-estimated camera poses from Structure-from-Motion algorithms, which frequently fail on in-the-wild videos. Second, using a single, global radiance field with finite representational capacity does not scale to longer trajectories in an unbounded scene. For handling unknown poses, we jointly estimate the camera poses with radiance field in a progressive manner. We show that progressive optimization significantly improves the robustness of the reconstruction. For handling large unbounded scenes, we dynamically allocate new local radiance fields trained with frames within a temporal window. This further improves robustness (e.g., performs well even under moderate pose drifts) and allows us to scale to large scenes. Our extensive evaluation on the Tanks and Temples dataset and our collected outdoor dataset, Static Hikes, show that our approach compares favorably with the state-of-the-art.
4:28

Progressively Optimized Local Radiance Fields for Robust View Synthesis paper page: present an algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video. The task poses two core challenges. First, most existing radiance field reconstruction approaches rely on accurate pre-estimated camera poses from Structure-from-Motion algorithms, which frequently fail on in-the-wild videos. Second, using a single, global radiance field with finite representational capacity does not scale to longer trajectories in an unbounded scene. For handling unknown poses, we jointly estimate the camera poses with radiance field in a progressive manner. We show that progressive optimization significantly improves the robustness of the reconstruction. For handling large unbounded scenes, we dynamically allocate new local radiance fields trained with frames within a temporal window. This further improves robustness (e.g., performs well even under moderate pose drifts) and allows us to scale to large scenes. Our extensive evaluation on the Tanks and Temples dataset and our collected outdoor dataset, Static Hikes, show that our approach compares favorably with the state-of-the-art.

AK

140,624 次观看 • 3 年前

VIDEO | Delhi: Professor Dr Aishwarya of Galgotias University responds over reports of the university being asked to vacate its stall at the AI Summit Expo, says, "No, we have not received any directive yet to vacate the expo area. In fact, we are ready to answer for what happened yesterday. A viral tweet circulated which went in a completely wrong direction and was misinterpreted. Everywhere, we have maintained the same claim—we never said that we manufacture or design these things. We have always said that our university is investing Rs 350 crore in AI. As part of that investment, we have brought different kinds of things into the university campus so that our students can explore, expand, and carry out research and development, and try to build an advanced prototype... actually it was misinterpreted. It is a very well-known fact, and we also know that this is something available everywhere on the internet. Why would we claim such a thing? It was misinterpreted, a viral tweet was posted and exaggerated—you know how such things happen. Therefore, we want to clarify that we are not a manufacturing unit. We are not people who make all these things. We are only here so that our university students can conduct research and development. As part of this investment, we are bringing these things in for that purpose.” #GalgotiasUniversity #IndiaAISummit2026 (Full video available on PTI Videos -
1:30

VIDEO | Delhi: Professor Dr Aishwarya of Galgotias University responds over reports of the university being asked to vacate its stall at the AI Summit Expo, says, "No, we have not received any directive yet to vacate the expo area. In fact, we are ready to answer for what happened yesterday. A viral tweet circulated which went in a completely wrong direction and was misinterpreted. Everywhere, we have maintained the same claim—we never said that we manufacture or design these things. We have always said that our university is investing Rs 350 crore in AI. As part of that investment, we have brought different kinds of things into the university campus so that our students can explore, expand, and carry out research and development, and try to build an advanced prototype... actually it was misinterpreted. It is a very well-known fact, and we also know that this is something available everywhere on the internet. Why would we claim such a thing? It was misinterpreted, a viral tweet was posted and exaggerated—you know how such things happen. Therefore, we want to clarify that we are not a manufacturing unit. We are not people who make all these things. We are only here so that our university students can conduct research and development. As part of this investment, we are bringing these things in for that purpose.” #GalgotiasUniversity #IndiaAISummit2026 (Full video available on PTI Videos -

Press Trust of India

83,384 次观看 • 4 个月前

Introducing RAGs v3 🌟: Build a RAG bot that can also search the web 🌐, to find answers that aren’t immediately in its corpus. Do this all in natural language, not code! 💬 Get a better experience than ChatGPT + Bing ⚡️ We used our integration with Metaphor - a search engine designed for LLMs - to pull in relevant text from the internet. Additionally: - you can now view the tools that the agent is using. - note: web search is only accessible for our OpenAI agent Check out the video below, and get started with the repo today. Repo:
1:17

Introducing RAGs v3 🌟: Build a RAG bot that can also search the web 🌐, to find answers that aren’t immediately in its corpus. Do this all in natural language, not code! 💬 Get a better experience than ChatGPT + Bing ⚡️ We used our integration with Metaphor - a search engine designed for LLMs - to pull in relevant text from the internet. Additionally: - you can now view the tools that the agent is using. - note: web search is only accessible for our OpenAI agent Check out the video below, and get started with the repo today. Repo:

LlamaIndex 🦙

76,366 次观看 • 2 年前

Just shipped scale orders on Pacifica Scale orders are limit orders that allow you to place multiple limit orders in a predetermined range. We made it a bit different: 1. You can drag the price ranges directly on the chart. 2. You can visualize the orders and avg price for that scale order before the orders are placed. 3. We also made scale orders very easy to manage. If you want to cancel a particular scale order - instead of having to cancel them one by one, you can click on open orders -> scale orders to cancel a particular scale order.
0:28

Just shipped scale orders on Pacifica Scale orders are limit orders that allow you to place multiple limit orders in a predetermined range. We made it a bit different: 1. You can drag the price ranges directly on the chart. 2. You can visualize the orders and avg price for that scale order before the orders are placed. 3. We also made scale orders very easy to manage. If you want to cancel a particular scale order - instead of having to cancel them one by one, you can click on open orders -> scale orders to cancel a particular scale order.

Jose

23,512 次观看 • 6 个月前

It's no wonder that so many young Americans have learned to hate their country. They were taught that we don't have a shared story and that America is divided. Our churches and social clubs have emptied out. Social media, which was advertised as a tool to bring us together, did the exact opposite. And they're soaking in hostile propaganda 8 hours a day from the phones in their pockets. Our kids are going to search for belonging no matter what we do. And there are plenty of voices out there offering a finished identity: 'THIS is what you believe. THIS is who YOU are.' So, what are we offering instead?
6:18

It's no wonder that so many young Americans have learned to hate their country. They were taught that we don't have a shared story and that America is divided. Our churches and social clubs have emptied out. Social media, which was advertised as a tool to bring us together, did the exact opposite. And they're soaking in hostile propaganda 8 hours a day from the phones in their pockets. Our kids are going to search for belonging no matter what we do. And there are plenty of voices out there offering a finished identity: 'THIS is what you believe. THIS is who YOU are.' So, what are we offering instead?

Glenn Beck

50,758 次观看 • 9 天前

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:
0:44

Today we announced our new Fairwater datacenter in Atlanta, connected with our first Fairwater site in Wisconsin and our broader Azure footprint to create the world’s first AI superfactory. Fairwater exemplifies our vision for a fungible fleet: infra that can serve any workload, anywhere, on fit-for-purpose accelerators and network paths, with maximum performance and efficiency. AI workloads have evolved beyond large-scale pre-training. Today, they encompass fine-tuning, reinforcement learning (RL), synthetic data generation, evaluation pipelines, and more. Fairwater is built to support this full lifecycle: Max density: Fairwater’s two-story design and liquid cooling system lets us place racks in three dimensions and pack them with GPUs as densely as possible, minimizing cable runs and improving latency and effective bandwidth. Fleet: Each Fairwater DC can integrate hundreds of thousands of the latest NVIDIA GPUs into a single coherent cluster. This provides flexible infra that can support the full spectrum of workloads, and ensure no GPU is left unnecessarily idle. And that’s on top of the more than 100,000 GB300s coming online this quarter alone for inference across the rest of our fleet. For us, it’s all about turning every gigawatt into the maximum number of useful tokens. Not every GW is created equal! Planet-scale: Every Fairwater DC will connect through our continent-spanning AI WAN to prior generations of AI supercomputers, forming a truly fungible pool of compute. This enables developers to scale beyond the capacity of a single site and dynamically land workloads on the right infra for their needs. Together, these innovations let us bring together different generations of silicon and AI systems across DCs and geos into a single elastic system that scales seamlessly across training and inference workloads And this elastic AI capacity is all available alongside all the other cloud services (compute, storage, databases, app services) that AI agents and workloads need. This is what we mean when we talk about building a fungible fleet – a single, unified platform that pushes the limits of performance per watt and per dollar. Read more:

Satya Nadella

907,060 次观看 • 7 个月前

Then came one of the biggest financial updates of the meeting...from Treasury Secretary Scott Bessent. What he revealed was staggering. Bessent said that America is now pulling in RECORD tariff revenues...between $500 BILLION and $1 TRILLION a year...and the budget deficit is already down 26% from Biden’s last year in office. These are not small numbers. Bessent explained how Trump flipped the system on its head. The days of unbalanced international trade are over, and with that comes a massive influx of cash directly affecting America’s deficit. “On the international front you have a leveled the international trading system whereby countries took advantage of us and that’s over. It’s OVER.” “The Treasury Department is in is taking in record tariff revenues that I had been saying was running at a rate of 300 billion a year.” “You chastised me for saying that—number is too low. And as usual, you we’re right, that we had a substantial, jump from July to August, and I think we’re going to see a bigger jump from August to September.” “So I think we could be on our way, well, over half a trillion, maybe the, towards the trillion dollar number.” “This administration, your administration, has made a meaningful dent in the budget deficit.” Even the CBO, which often clashes with the Trump administration, had to concede: the deficit is projected to drop by $4 trillion over the next decade. “The average budget deficit during this term is 26% less than the last 12 months under Biden and even the CBO, and we don’t agree with CBO on everything as you said, last Friday—had to admit that they believe over the next ten years the budget deficit will be 4 trillion lower than they had previously scored.” And according to Bessent, that number could go even higher. “4 trillion: 3.3 trillion of tariff income, 700 billion of lower interest cost, and, you know, I would expect that that number could go up from here.”
1:34

Then came one of the biggest financial updates of the meeting...from Treasury Secretary Scott Bessent. What he revealed was staggering. Bessent said that America is now pulling in RECORD tariff revenues...between $500 BILLION and $1 TRILLION a year...and the budget deficit is already down 26% from Biden’s last year in office. These are not small numbers. Bessent explained how Trump flipped the system on its head. The days of unbalanced international trade are over, and with that comes a massive influx of cash directly affecting America’s deficit. “On the international front you have a leveled the international trading system whereby countries took advantage of us and that’s over. It’s OVER.” “The Treasury Department is in is taking in record tariff revenues that I had been saying was running at a rate of 300 billion a year.” “You chastised me for saying that—number is too low. And as usual, you we’re right, that we had a substantial, jump from July to August, and I think we’re going to see a bigger jump from August to September.” “So I think we could be on our way, well, over half a trillion, maybe the, towards the trillion dollar number.” “This administration, your administration, has made a meaningful dent in the budget deficit.” Even the CBO, which often clashes with the Trump administration, had to concede: the deficit is projected to drop by $4 trillion over the next decade. “The average budget deficit during this term is 26% less than the last 12 months under Biden and even the CBO, and we don’t agree with CBO on everything as you said, last Friday—had to admit that they believe over the next ten years the budget deficit will be 4 trillion lower than they had previously scored.” And according to Bessent, that number could go even higher. “4 trillion: 3.3 trillion of tariff income, 700 billion of lower interest cost, and, you know, I would expect that that number could go up from here.”

The Vigilant Fox 🦊

238,581 次观看 • 9 个月前