Nvidia Reveals Blackwell B200 GPU, The "World’s Most Powerful Chip" for AI

The Verge03-19

【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...

Source Link

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Report

Comment（3）

Kong123
·03-19
Strength and big potential of growth.... uninterrupted positive energy
Reply
2
Report
Kong123
·03-19
Wanderful, good job
Reply
1
Report
donaldong
·03-19
Amazing advancement in the AI industry. The 4th Indistrial Revolutionization is happening, and the scale and speed of this revolution will be unprecedented! Largest tech companies are desperate to adopt and drive forward. Those lacking foresight will be dissolved.
Reply
Like
Report

Most Discussed

7x24

All

Terms and Conditions·Privacy Policy

Company: TTMF Limited. Tech supported by Xiangshang Yixin.

Email：uservice@ttm.financial

{"i18n":{"language":"en_US"},"data":{"share":"https://ttm.financial/m/news/2420238462?lang=en_US&edition=fundamental","thumbnail":"","is_english":true,"pubTime":"2024-03-19 06:40","share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","id":"2420238462","market":"us","top_or_hot":-1,"title":"Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI","media":"The Verge","content":"<div>\n<p>【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...</p>\n\n<a href=\"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai\">Source Link</a>\n\n</div>\n","source":"theverge_highlight","html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nNvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI\n</h2>\n\n<h4 class=\"meta\">\n\n\n2024-03-19 06:40 GMT+8&nbsp;&nbsp;&nbsp;<a href=https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai><strong>The Verge</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...</p>\n\n<a href=\"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai\">Source Link</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","isBrief":false,"type":0,"news_type":1,"symbol":"BK4538","symbol_name":"云计算","start_time":0,"source_url":"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai","article_id":"2420238462","we_media_id":null,"thumbnails":[],"rights":null,"url":"https://stock-news.laohu8.com/highlight/detail?id=2420238462","pubTimestamp":1710801655,"sourceInfo":{"source_id":"theverge_highlight","name":"The Verge"},"weMediaInfo":null,"summary":"Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about","collect":0,"end_time":0,"defaultTopTitle":"theverge.com","property":[],"viewcount":null,"language":"en","relate_stocks":{"BK4538":"云计算","BK4543":"AI","LU0308772762.SGD":"Blackrock Global Allocation A2 SGD-H","LU0354030438.USD":"富国美国大盘成长基金Cl A Acc","LU0965509101.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"A\" (SGDHDG) ACC","IE00B7KXQ091.USD":"Janus Henderson Balanced A Inc USD","IE00B1BXHZ80.USD":"Legg Mason ClearBridge - US Appreciation A Acc USD","BK4533":"AQR资本管理(全球第二大对冲基金)","LU0494093205.USD":"贝莱德ESG灵活多元资产A2 USD-H","LU0417517546.SGD":"Allianz US Equity Cl AT Acc SGD","BK4534":"瑞士信贷持仓","BK4567":"ESG概念","LU0256863811.USD":"ALLIANZ US EQUITY \"A\" INC","LU0289960550.SGD":"AB FCP I - GLOBAL EQUITY BLEND PORTFOLIO 'A' (SGD) ACC","BK4553":"喜马拉雅资本持仓","LU1804176565.USD":"EASTSPRING INV GLOBAL GROWTH EQUITY \"A\" (USD) ACC","LU0310799852.SGD":"FTIF - Templeton Global Equity Income A MDIS SGD","BK4554":"元宇宙及AR概念","BK4532":"文艺复兴科技持仓","LU1814569148.SGD":"WELLINGTON GLOBAL QUALITY GROWTH \"D\" (SGDHDG) ACC","BK4514":"搜索引擎","LU0312595415.SGD":"Schroder ISF Global Climate Change Equity A Acc SGD","IE00B19Z9505.USD":"美盛-美国大盘成长股A Acc","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","LU0348723411.USD":"ALLIANZ GLOBAL HI-TECH GROWTH \"A\" (USD) INC","BK4548":"巴美列捷福持仓","LU0353189680.USD":"富国美国全盘成长基金Cl A Acc","IE00BJTD4N35.SGD":"Neuberger Berman US Long Short Equity A1  Acc SGD-H","LU0211327993.USD":"TEMPLETON GLOBAL EQUITY INCOME \"A\" (USD) ACC","IE00B3S45H60.SGD":"Neuberger Berman US Multicap Opportunities A Acc SGD-H","BK4549":"软银资本持仓","IE00BFSS8Q28.SGD":"Janus Henderson Balanced A Inc SGD-H","LU0109391861.USD":"富兰克林美国机遇基金A Acc","LU1280957306.USD":"THREADNEEDLE (LUX) US CONTRARIAN CORE EQUITIES \"AUP\" (USD) INC","LU0061474705.USD":"THREADNEEDLE (LUX) GLOBAL DYNAMIC REAL RETURN \"AU\" (USD) ACC","LU0316494557.USD":"FRANKLIN GLOBAL FUNDAMENTAL STRATEGIES \"A\" ACC","IE00BMPRXR70.SGD":"Neuberger Berman 5G Connectivity A Acc SGD-H","LU0079474960.USD":"联博美国增长基金A","IE00BFSS7M15.SGD":"Janus Henderson Balanced A Acc SGD-H","IE00BKDWB100.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5H\" (SGDHDG) ACC","LU0354030511.USD":"ALLSPRING  U.S. LARGE CAP GROWTH \"I\" (USD) ACC","LU1691799644.USD":"Amundi Funds Polen Capital Global Growth A2 (C) USD","LU0965509283.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (SGDHDG) INC","LU1914381329.SGD":"Allianz Best Styles Global Equity Cl ET Acc H2-SGD","LU0061474960.USD":"天利环球焦点基金AU Acc","BK4550":"红杉资本持仓","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD"},"translate_title":"英伟达揭示了布莱克威尔B200 GPU，人工智能的“世界上最强大的芯片”","themeId":"","isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{},"content_text":"【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.Nvidia shares slipped 1.4% in after hours trading.Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.Here’s what one GB200 looks like. Two GPUs, one CPU, one board.Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.Nvidia says it’s adding both FP4 and FP6 with Blackwell.Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 exaflops) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.The GB200 NVL72.Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.","kind":null,"is_publish_news":true,"is_publish_highlight":true,"is_publish_live":false,"is_publish_wemedia":null,"editions":null,"column":"","sentiment":"0","news_tag":"","news_rank":0,"symbols":[],"gpt_button":0},"commentList":[{"id":285925450744088,"gmtCreate":1710840819211,"gmtModify":1710840822728,"author":{"id":"3585634044713216","authorId":"3585634044713216","name":"Kong123","avatar":"https://community-static.tradeup.com/news/b98348643175392a7296a947f9f8739a","vip":1,"hat":"https://static.tigerbbs.com/0a8758340a08196056f8bc1e54805193","crmLevel":8,"crmLevelSwitch":0,"idStr":"3585634044713216","authorIdStr":"3585634044713216"},"htmlText":"Strength and big potential of growth.... uninterrupted positive energy","listText":"Strength and big potential of growth.... uninterrupted positive energy","text":"Strength and big potential of growth.... uninterrupted positive energy","images":[],"top":1,"highlighted":1,"essential":1,"paper":1,"likeSize":2,"commentSize":0,"repostSize":0,"link":"https://ttm.financial/post/285925450744088","repostId":2420238462,"repostType":2,"repost":{"id":"2420238462","pubTimestamp":1710801655,"share":"https://ttm.financial/m/news/2420238462?lang=&edition=fundamental","pubTime":"2024-03-19 06:40","market":"us","language":"en","title":"Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI","url":"https://stock-news.laohu8.com/highlight/detail?id=2420238462","media":"The Verge","summary":"Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about","content":"<html><head></head><body><p>【<a href=\"https://ttm.financial/RN?name=RNLive&amp;rndata=%7B%22liveId%22:%221793027278299171%22,%22type%22:1%7D\" title=\"Watch Now: GTC Jensen Huang’s Keynote Speech\" target=\"_blank\">Watch Now: GTC Jensen Huang’s Keynote Speech</a>】</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/881ba52cdbeba404e9f9193b8b982753\" title=\"The Blackwell B200 GPU.\" tg-width=\"2400\" tg-height=\"1600\"/><span>The Blackwell B200 GPU.</span></p><p>Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/faa3f418929a50c10b1a17ba95e16329\" title=\"Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.\" tg-width=\"2400\" tg-height=\"1425\"/><span>Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.</span></p><p>Nvidia shares slipped 1.4% in after hours trading.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/48dc2c9766cac9fe87e0b25e3dfeadf5\" title=\"\" tg-width=\"847\" tg-height=\"829\"/></p><p>Nvidia says the new B200 GPU offers up to 20 <em>petaflops </em>of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.</p><p>On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9b2dc6bf1fce0939104ea042dbcc96ee\" title=\"Here’s what one GB200 looks like. Two GPUs, one CPU, one board.\" tg-width=\"2400\" tg-height=\"1633\"/><span>Here’s what one GB200 looks like. Two GPUs, one CPU, one board.</span></p><p>Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.</p><p>That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9147c538f644bcdae0bd7421a57fa706\" title=\"Nvidia says it’s adding both FP4 and FP6 with Blackwell.\" tg-width=\"2400\" tg-height=\"1363\"/><span>Nvidia says it’s adding both FP4 and FP6 with Blackwell.</span></p><p>Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.</p><p>Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 <em>exaflops</em>) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/3c31b5f312fa568f1ca0881310b23248\" title=\"The GB200 NVL72.\" tg-width=\"2400\" tg-height=\"1814\"/><span>The GB200 NVL72.</span></p><p>Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.</p><p>The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.</p><p>And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/5fed30ea06faa08938ddcdf171f7594c\" title=\"\" tg-width=\"2400\" tg-height=\"962\"/></p><p>Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).</p><p>We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.</p></body></html>","source":"theverge_highlight","collect":0,"html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nNvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI\n</h2>\n\n<h4 class=\"meta\">\n\n\n2024-03-19 06:40 GMT+8&nbsp;&nbsp;&nbsp;<a href=https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai><strong>The Verge</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...</p>\n\n<a href=\"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai\">Web Link</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","type":0,"thumbnail":"","relate_stocks":{"IE00B3S45H60.SGD":"Neuberger Berman US Multicap Opportunities A Acc SGD-H","BK4550":"红杉资本持仓","LU0109391861.USD":"富兰克林美国机遇基金A Acc","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD","IE00B19Z9505.USD":"美盛-美国大盘成长股A Acc","LU0417517546.SGD":"Allianz US Equity Cl AT Acc SGD","LU0079474960.USD":"联博美国增长基金A","LU0312595415.SGD":"Schroder ISF Global Climate Change Equity A Acc SGD","LU0353189680.USD":"富国美国全盘成长基金Cl A Acc","LU0061474960.USD":"天利环球焦点基金AU Acc","LU1280957306.USD":"THREADNEEDLE (LUX) US CONTRARIAN CORE EQUITIES \"AUP\" (USD) INC","LU0308772762.SGD":"Blackrock Global Allocation A2 SGD-H","IE00BMPRXR70.SGD":"Neuberger Berman 5G Connectivity A Acc SGD-H","LU0494093205.USD":"贝莱德ESG灵活多元资产A2 USD-H","IE00BKDWB100.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5H\" (SGDHDG) ACC","LU1914381329.SGD":"Allianz Best Styles Global Equity Cl ET Acc H2-SGD","BK4549":"软银资本持仓","LU0310799852.SGD":"FTIF - Templeton Global Equity Income A MDIS SGD","LU1804176565.USD":"EASTSPRING INV GLOBAL GROWTH EQUITY \"A\" (USD) ACC","BK4514":"搜索引擎","BK4548":"巴美列捷福持仓","BK4538":"云计算","LU0061474705.USD":"THREADNEEDLE (LUX) GLOBAL DYNAMIC REAL RETURN \"AU\" (USD) ACC","BK4554":"元宇宙及AR概念","LU1814569148.SGD":"WELLINGTON GLOBAL QUALITY GROWTH \"D\" (SGDHDG) ACC","BK4532":"文艺复兴科技持仓","LU1691799644.USD":"Amundi Funds Polen Capital Global Growth A2 (C) USD","IE00B1BXHZ80.USD":"Legg Mason ClearBridge - US Appreciation A Acc USD","LU0289960550.SGD":"AB FCP I - GLOBAL EQUITY BLEND PORTFOLIO 'A' (SGD) ACC","LU0965509283.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (SGDHDG) INC","LU0316494557.USD":"FRANKLIN GLOBAL FUNDAMENTAL STRATEGIES \"A\" ACC","BK4553":"喜马拉雅资本持仓","BK4567":"ESG概念","BK4534":"瑞士信贷持仓","IE00BFSS8Q28.SGD":"Janus Henderson Balanced A Inc SGD-H","LU0211327993.USD":"TEMPLETON GLOBAL EQUITY INCOME \"A\" (USD) ACC","LU0354030511.USD":"ALLSPRING  U.S. LARGE CAP GROWTH \"I\" (USD) ACC","BK4533":"AQR资本管理(全球第二大对冲基金)","LU0354030438.USD":"富国美国大盘成长基金Cl A Acc","LU0256863811.USD":"ALLIANZ US EQUITY \"A\" INC","IE00BJTD4N35.SGD":"Neuberger Berman US Long Short Equity A1  Acc SGD-H","LU0965509101.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"A\" (SGDHDG) ACC","IE00B7KXQ091.USD":"Janus Henderson Balanced A Inc USD","LU0348723411.USD":"ALLIANZ GLOBAL HI-TECH GROWTH \"A\" (USD) INC","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","BK4543":"AI","IE00BFSS7M15.SGD":"Janus Henderson Balanced A Acc SGD-H"},"source_url":"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai","is_english":true,"share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","article_id":"2420238462","content_text":"【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.Nvidia shares slipped 1.4% in after hours trading.Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.Here’s what one GB200 looks like. Two GPUs, one CPU, one board.Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.Nvidia says it’s adding both FP4 and FP6 with Blackwell.Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 exaflops) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.The GB200 NVL72.Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.","news_type":1},"isVote":1,"likeStatus":false,"favoriteStatus":false,"reportStatus":false,"tweetType":1,"langContent":"EN"},{"id":285924617486496,"gmtCreate":1710840631338,"gmtModify":1710840634814,"author":{"id":"3585634044713216","authorId":"3585634044713216","name":"Kong123","avatar":"https://community-static.tradeup.com/news/b98348643175392a7296a947f9f8739a","vip":1,"hat":"https://static.tigerbbs.com/0a8758340a08196056f8bc1e54805193","crmLevel":8,"crmLevelSwitch":0,"idStr":"3585634044713216","authorIdStr":"3585634044713216"},"htmlText":"Wanderful, good job","listText":"Wanderful, good job","text":"Wanderful, good job","images":[],"top":1,"highlighted":1,"essential":1,"paper":1,"likeSize":1,"commentSize":0,"repostSize":0,"link":"https://ttm.financial/post/285924617486496","repostId":2420238462,"repostType":2,"repost":{"id":"2420238462","pubTimestamp":1710801655,"share":"https://ttm.financial/m/news/2420238462?lang=&edition=fundamental","pubTime":"2024-03-19 06:40","market":"us","language":"en","title":"Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI","url":"https://stock-news.laohu8.com/highlight/detail?id=2420238462","media":"The Verge","summary":"Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about","content":"<html><head></head><body><p>【<a href=\"https://ttm.financial/RN?name=RNLive&amp;rndata=%7B%22liveId%22:%221793027278299171%22,%22type%22:1%7D\" title=\"Watch Now: GTC Jensen Huang’s Keynote Speech\" target=\"_blank\">Watch Now: GTC Jensen Huang’s Keynote Speech</a>】</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/881ba52cdbeba404e9f9193b8b982753\" title=\"The Blackwell B200 GPU.\" tg-width=\"2400\" tg-height=\"1600\"/><span>The Blackwell B200 GPU.</span></p><p>Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/faa3f418929a50c10b1a17ba95e16329\" title=\"Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.\" tg-width=\"2400\" tg-height=\"1425\"/><span>Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.</span></p><p>Nvidia shares slipped 1.4% in after hours trading.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/48dc2c9766cac9fe87e0b25e3dfeadf5\" title=\"\" tg-width=\"847\" tg-height=\"829\"/></p><p>Nvidia says the new B200 GPU offers up to 20 <em>petaflops </em>of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.</p><p>On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9b2dc6bf1fce0939104ea042dbcc96ee\" title=\"Here’s what one GB200 looks like. Two GPUs, one CPU, one board.\" tg-width=\"2400\" tg-height=\"1633\"/><span>Here’s what one GB200 looks like. Two GPUs, one CPU, one board.</span></p><p>Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.</p><p>That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9147c538f644bcdae0bd7421a57fa706\" title=\"Nvidia says it’s adding both FP4 and FP6 with Blackwell.\" tg-width=\"2400\" tg-height=\"1363\"/><span>Nvidia says it’s adding both FP4 and FP6 with Blackwell.</span></p><p>Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.</p><p>Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 <em>exaflops</em>) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/3c31b5f312fa568f1ca0881310b23248\" title=\"The GB200 NVL72.\" tg-width=\"2400\" tg-height=\"1814\"/><span>The GB200 NVL72.</span></p><p>Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.</p><p>The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.</p><p>And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/5fed30ea06faa08938ddcdf171f7594c\" title=\"\" tg-width=\"2400\" tg-height=\"962\"/></p><p>Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).</p><p>We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.</p></body></html>","source":"theverge_highlight","collect":0,"html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nNvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI\n</h2>\n\n<h4 class=\"meta\">\n\n\n2024-03-19 06:40 GMT+8&nbsp;&nbsp;&nbsp;<a href=https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai><strong>The Verge</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...</p>\n\n<a href=\"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai\">Web Link</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","type":0,"thumbnail":"","relate_stocks":{"IE00B3S45H60.SGD":"Neuberger Berman US Multicap Opportunities A Acc SGD-H","BK4550":"红杉资本持仓","LU0109391861.USD":"富兰克林美国机遇基金A Acc","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD","IE00B19Z9505.USD":"美盛-美国大盘成长股A Acc","LU0417517546.SGD":"Allianz US Equity Cl AT Acc SGD","LU0079474960.USD":"联博美国增长基金A","LU0312595415.SGD":"Schroder ISF Global Climate Change Equity A Acc SGD","LU0353189680.USD":"富国美国全盘成长基金Cl A Acc","LU0061474960.USD":"天利环球焦点基金AU Acc","LU1280957306.USD":"THREADNEEDLE (LUX) US CONTRARIAN CORE EQUITIES \"AUP\" (USD) INC","LU0308772762.SGD":"Blackrock Global Allocation A2 SGD-H","IE00BMPRXR70.SGD":"Neuberger Berman 5G Connectivity A Acc SGD-H","LU0494093205.USD":"贝莱德ESG灵活多元资产A2 USD-H","IE00BKDWB100.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5H\" (SGDHDG) ACC","LU1914381329.SGD":"Allianz Best Styles Global Equity Cl ET Acc H2-SGD","BK4549":"软银资本持仓","LU0310799852.SGD":"FTIF - Templeton Global Equity Income A MDIS SGD","LU1804176565.USD":"EASTSPRING INV GLOBAL GROWTH EQUITY \"A\" (USD) ACC","BK4514":"搜索引擎","BK4548":"巴美列捷福持仓","BK4538":"云计算","LU0061474705.USD":"THREADNEEDLE (LUX) GLOBAL DYNAMIC REAL RETURN \"AU\" (USD) ACC","BK4554":"元宇宙及AR概念","LU1814569148.SGD":"WELLINGTON GLOBAL QUALITY GROWTH \"D\" (SGDHDG) ACC","BK4532":"文艺复兴科技持仓","LU1691799644.USD":"Amundi Funds Polen Capital Global Growth A2 (C) USD","IE00B1BXHZ80.USD":"Legg Mason ClearBridge - US Appreciation A Acc USD","LU0289960550.SGD":"AB FCP I - GLOBAL EQUITY BLEND PORTFOLIO 'A' (SGD) ACC","LU0965509283.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (SGDHDG) INC","LU0316494557.USD":"FRANKLIN GLOBAL FUNDAMENTAL STRATEGIES \"A\" ACC","BK4553":"喜马拉雅资本持仓","BK4567":"ESG概念","BK4534":"瑞士信贷持仓","IE00BFSS8Q28.SGD":"Janus Henderson Balanced A Inc SGD-H","LU0211327993.USD":"TEMPLETON GLOBAL EQUITY INCOME \"A\" (USD) ACC","LU0354030511.USD":"ALLSPRING  U.S. LARGE CAP GROWTH \"I\" (USD) ACC","BK4533":"AQR资本管理(全球第二大对冲基金)","LU0354030438.USD":"富国美国大盘成长基金Cl A Acc","LU0256863811.USD":"ALLIANZ US EQUITY \"A\" INC","IE00BJTD4N35.SGD":"Neuberger Berman US Long Short Equity A1  Acc SGD-H","LU0965509101.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"A\" (SGDHDG) ACC","IE00B7KXQ091.USD":"Janus Henderson Balanced A Inc USD","LU0348723411.USD":"ALLIANZ GLOBAL HI-TECH GROWTH \"A\" (USD) INC","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","BK4543":"AI","IE00BFSS7M15.SGD":"Janus Henderson Balanced A Acc SGD-H"},"source_url":"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai","is_english":true,"share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","article_id":"2420238462","content_text":"【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.Nvidia shares slipped 1.4% in after hours trading.Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.Here’s what one GB200 looks like. Two GPUs, one CPU, one board.Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.Nvidia says it’s adding both FP4 and FP6 with Blackwell.Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 exaflops) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.The GB200 NVL72.Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.","news_type":1},"isVote":1,"likeStatus":false,"favoriteStatus":false,"reportStatus":false,"tweetType":1,"langContent":"EN"},{"id":285849616838832,"gmtCreate":1710822211589,"gmtModify":1710823660044,"author":{"id":"4089867579726070","authorId":"4089867579726070","name":"donaldong","avatar":"https://static.tigerbbs.com/d64f17b77d1eff1405fb753b45503592","vip":1,"crmLevel":6,"crmLevelSwitch":1,"idStr":"4089867579726070","authorIdStr":"4089867579726070"},"htmlText":"Amazing advancement in the AI industry. The 4th Indistrial Revolutionization is happening, and the scale and speed of this revolution will be unprecedented! Largest tech companies are desperate to adopt and drive forward. Those lacking foresight will be dissolved.&nbsp;","listText":"Amazing advancement in the AI industry. The 4th Indistrial Revolutionization is happening, and the scale and speed of this revolution will be unprecedented! Largest tech companies are desperate to adopt and drive forward. Those lacking foresight will be dissolved.&nbsp;","text":"Amazing advancement in the AI industry. The 4th Indistrial Revolutionization is happening, and the scale and speed of this revolution will be unprecedented! Largest tech companies are desperate to adopt and drive forward. Those lacking foresight will be dissolved.","images":[],"top":1,"highlighted":1,"essential":1,"paper":1,"likeSize":0,"commentSize":0,"repostSize":0,"link":"https://ttm.financial/post/285849616838832","repostId":2420238462,"repostType":2,"repost":{"id":"2420238462","pubTimestamp":1710801655,"share":"https://ttm.financial/m/news/2420238462?lang=&edition=fundamental","pubTime":"2024-03-19 06:40","market":"us","language":"en","title":"Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI","url":"https://stock-news.laohu8.com/highlight/detail?id=2420238462","media":"The Verge","summary":"Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about","content":"<html><head></head><body><p>【<a href=\"https://ttm.financial/RN?name=RNLive&amp;rndata=%7B%22liveId%22:%221793027278299171%22,%22type%22:1%7D\" title=\"Watch Now: GTC Jensen Huang’s Keynote Speech\" target=\"_blank\">Watch Now: GTC Jensen Huang’s Keynote Speech</a>】</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/881ba52cdbeba404e9f9193b8b982753\" title=\"The Blackwell B200 GPU.\" tg-width=\"2400\" tg-height=\"1600\"/><span>The Blackwell B200 GPU.</span></p><p>Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/faa3f418929a50c10b1a17ba95e16329\" title=\"Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.\" tg-width=\"2400\" tg-height=\"1425\"/><span>Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.</span></p><p>Nvidia shares slipped 1.4% in after hours trading.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/48dc2c9766cac9fe87e0b25e3dfeadf5\" title=\"\" tg-width=\"847\" tg-height=\"829\"/></p><p>Nvidia says the new B200 GPU offers up to 20 <em>petaflops </em>of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.</p><p>On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9b2dc6bf1fce0939104ea042dbcc96ee\" title=\"Here’s what one GB200 looks like. Two GPUs, one CPU, one board.\" tg-width=\"2400\" tg-height=\"1633\"/><span>Here’s what one GB200 looks like. Two GPUs, one CPU, one board.</span></p><p>Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.</p><p>That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/9147c538f644bcdae0bd7421a57fa706\" title=\"Nvidia says it’s adding both FP4 and FP6 with Blackwell.\" tg-width=\"2400\" tg-height=\"1363\"/><span>Nvidia says it’s adding both FP4 and FP6 with Blackwell.</span></p><p>Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.</p><p>Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 <em>exaflops</em>) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/3c31b5f312fa568f1ca0881310b23248\" title=\"The GB200 NVL72.\" tg-width=\"2400\" tg-height=\"1814\"/><span>The GB200 NVL72.</span></p><p>Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.</p><p>The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.</p><p>And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.</p><p class=\"t-img-caption\"><img src=\"https://community-static.tradeup.com/news/5fed30ea06faa08938ddcdf171f7594c\" title=\"\" tg-width=\"2400\" tg-height=\"962\"/></p><p>Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).</p><p>We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.</p></body></html>","source":"theverge_highlight","collect":0,"html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>Nvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\nNvidia Reveals Blackwell B200 GPU, The \"World’s Most Powerful Chip\" for AI\n</h2>\n\n<h4 class=\"meta\">\n\n\n2024-03-19 06:40 GMT+8&nbsp;&nbsp;&nbsp;<a href=https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai><strong>The Verge</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and ...</p>\n\n<a href=\"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai\">Web Link</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","type":0,"thumbnail":"","relate_stocks":{"IE00B3S45H60.SGD":"Neuberger Berman US Multicap Opportunities A Acc SGD-H","BK4550":"红杉资本持仓","LU0109391861.USD":"富兰克林美国机遇基金A Acc","IE00BJJMRX11.SGD":"Janus Henderson Balanced A Acc SGD","IE00B19Z9505.USD":"美盛-美国大盘成长股A Acc","LU0417517546.SGD":"Allianz US Equity Cl AT Acc SGD","LU0079474960.USD":"联博美国增长基金A","LU0312595415.SGD":"Schroder ISF Global Climate Change Equity A Acc SGD","LU0353189680.USD":"富国美国全盘成长基金Cl A Acc","LU0061474960.USD":"天利环球焦点基金AU Acc","LU1280957306.USD":"THREADNEEDLE (LUX) US CONTRARIAN CORE EQUITIES \"AUP\" (USD) INC","LU0308772762.SGD":"Blackrock Global Allocation A2 SGD-H","IE00BMPRXR70.SGD":"Neuberger Berman 5G Connectivity A Acc SGD-H","LU0494093205.USD":"贝莱德ESG灵活多元资产A2 USD-H","IE00BKDWB100.SGD":"PINEBRIDGE US LARGE CAP RESEARCH ENHANCED \"A5H\" (SGDHDG) ACC","LU1914381329.SGD":"Allianz Best Styles Global Equity Cl ET Acc H2-SGD","BK4549":"软银资本持仓","LU0310799852.SGD":"FTIF - Templeton Global Equity Income A MDIS SGD","LU1804176565.USD":"EASTSPRING INV GLOBAL GROWTH EQUITY \"A\" (USD) ACC","BK4514":"搜索引擎","BK4548":"巴美列捷福持仓","BK4538":"云计算","LU0061474705.USD":"THREADNEEDLE (LUX) GLOBAL DYNAMIC REAL RETURN \"AU\" (USD) ACC","BK4554":"元宇宙及AR概念","LU1814569148.SGD":"WELLINGTON GLOBAL QUALITY GROWTH \"D\" (SGDHDG) ACC","BK4532":"文艺复兴科技持仓","LU1691799644.USD":"Amundi Funds Polen Capital Global Growth A2 (C) USD","IE00B1BXHZ80.USD":"Legg Mason ClearBridge - US Appreciation A Acc USD","LU0289960550.SGD":"AB FCP I - GLOBAL EQUITY BLEND PORTFOLIO 'A' (SGD) ACC","LU0965509283.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"AD\" (SGDHDG) INC","LU0316494557.USD":"FRANKLIN GLOBAL FUNDAMENTAL STRATEGIES \"A\" ACC","BK4553":"喜马拉雅资本持仓","BK4567":"ESG概念","BK4534":"瑞士信贷持仓","IE00BFSS8Q28.SGD":"Janus Henderson Balanced A Inc SGD-H","LU0211327993.USD":"TEMPLETON GLOBAL EQUITY INCOME \"A\" (USD) ACC","LU0354030511.USD":"ALLSPRING  U.S. LARGE CAP GROWTH \"I\" (USD) ACC","BK4533":"AQR资本管理(全球第二大对冲基金)","LU0354030438.USD":"富国美国大盘成长基金Cl A Acc","LU0256863811.USD":"ALLIANZ US EQUITY \"A\" INC","IE00BJTD4N35.SGD":"Neuberger Berman US Long Short Equity A1  Acc SGD-H","LU0965509101.SGD":"AB LOW VOLATILITY EQUITY PORTFOLIO \"A\" (SGDHDG) ACC","IE00B7KXQ091.USD":"Janus Henderson Balanced A Inc USD","LU0348723411.USD":"ALLIANZ GLOBAL HI-TECH GROWTH \"A\" (USD) INC","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","BK4543":"AI","IE00BFSS7M15.SGD":"Janus Henderson Balanced A Acc SGD-H"},"source_url":"https://www.theverge.com/2024/3/18/24105157/nvidia-blackwell-gpu-b200-ai","is_english":true,"share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","article_id":"2420238462","content_text":"【Watch Now: GTC Jensen Huang’s Keynote Speech】The Blackwell B200 GPU.Nvidia’s must-have H100 AI chip made it a multitrillion-dollar company, one that may be worth more than Alphabet and Amazon, and competitors have been fighting to catch up. But perhaps Nvidia is about to extend its lead — with the new Blackwell B200 GPU and GB200 “superchip.”Nvidia CEO Jensen Huang holds up his new GPU on the left, next to an H100 on the right, from the GTC livestream.Nvidia shares slipped 1.4% in after hours trading.Nvidia says the new B200 GPU offers up to 20 petaflops of FP4 horsepower from its 208 billion transistors, and that a GB200 that combines two of those GPUs with a single Grace CPU can offer 30 times the performance for LLM inference workloads while also potentially being substantially more efficient. It “reduces cost and energy consumption by up to 25x” over an H100, says Nvidia.On a GPT-3 LLM benchmark with 175 billion parameters, Nvidia says the GB200 has a somewhat more modest 7 times the performance of an H100, and Nvidia says it offers 4x the training speed.Here’s what one GB200 looks like. Two GPUs, one CPU, one board.Nvidia told journalists one of the key differences is a second-gen transformer engine that doubles the compute, bandwidth, and model size by using four bits for each neuron instead of eight (thus the 20 petaflops of FP4 I mention earlier.) A second key difference only comes when you link up huge numbers of these GPUs in a server: a next-gen NVLink networking solution that lets 576 GPUs talk to each other, with 1.8 terabytes per second of bidirectional bandwidth.That required Nvidia to build an entire new network switch chip, one with 50 billion transistors and some of its own onboard compute: 3.6 teraflops of FP8, says Nvidia.Nvidia says it’s adding both FP4 and FP6 with Blackwell.Previously, Nvidia says, a cluster of just 16 GPUs would spend 60 percent of their time communicating with one another and only 40 percent actually computing.Nvidia is counting on companies buying large quantities of these GPUs, of course, and is packaging them in larger supercomputer-ready designs, like the GB200 NVL72 which plugs 36 CPUs and 72 GPUs into a single liquid-cooled rack for a total of 720 petaflops of AI training performance or 1,440 petaflops (aka 1.4 exaflops) of inference. It has nearly two miles of cables inside, with 5,000 individual cables.The GB200 NVL72.Each tray in the rack contains either two GB200 chips, or two NVLink switches, with 18 of the former and 9 of the latter per rack. In total, Nvidia says one of these racks can support a 27-trillion parameter model. GPT-4 is rumored to be around a 1.7-trillion parameter model.The company says Amazon, Google, Microsoft, and Oracle are all already planning to offer the NVL72 racks in their cloud service offerings, though it’s not clear how many they’re buying.And of course, Nvidia is happy to offer companies the rest of the solution, too. Here’s the DGX Superpod for DGX GB200, which combines eight systems in one for a total of 288 CPUs, 576 GPUs, 240TB of memory, and 11.5 exaflops of FP4 computing.Nvidia says its systems can scale to tens of thousands of the GB200 superchips, connected together with 800Gbps networking with its new Quantum-X800 Infiniband (for up to 144 connections) or Spectrum-X800 Ethernet (for up to 64 connections).We don’t expect to hear anything about new gaming GPUs today, as this news is coming out of Nvidia’s GPU Technology Conference, which is usually almost entirely focused on GPU computing and AI, not gaming. But the Blackwell GPU architecture will likely also power a future RTX 50-series lineup of desktop graphics cards.","news_type":1},"isVote":1,"likeStatus":false,"favoriteStatus":false,"reportStatus":false,"tweetType":1,"langContent":"EN"}],"hasMoreComments":true,"newsSizeData":{"likeSize":16,"commentSize":3,"repostSize":4,"favoriteSize":1,"likeStatus":false,"favoriteStatus":false}}