AMD 联合 IBM 与 Zyphra 发布全球首款全 AMD 硬件训练 MoE 模型 ZAYA1

牛透社11-26

牛透社 11 月 25 日消息，AMD 联合 IBM 与人工智能初创公司 Zyphra 共同发布全球首个全 AMD 硬件训练的 MoE 基础模型 ZAYA1。该模型预训练数据量达 14 万亿 token，综合性能与 Qwen3 系列持平，在数学/STEM 推理任务中无需指令微调即可接近 Qwen3 专业版水平。该模型基于 IBM Cloud 提供的 128 个计算节点（总计 1024 张 AMD ...

网页链接

免责声明：本文观点仅代表作者个人观点，不构成本平台的投资建议，本平台不对文章信息准确性、完整性和及时性做出任何保证，亦不对因使用或信赖文章信息引发的任何损失承担责任。

精彩评论

我们需要你的真知灼见来填补这片空白

发表看法

{"i18n":{"language":"zh_CN"},"isChannel":false,"data":{"share":"https://www.laohu8.com/m/news/2586404116?lang=zh_CN&edition=full","thumbnail":"","is_english":false,"pubTime":"2025-11-26 05:55","share_image_url":"https://static.laohu8.com/9a95c1376e76363c1401fee7d3717173","id":"2586404116","market":"us","top_or_hot":-1,"title":"AMD 联合 IBM 与 Zyphra 发布全球首款全 AMD 硬件训练 MoE 模型 ZAYA1","media":"牛透社","content":"<div>\n<p>牛透社 11 月 25 日消息，AMD 联合 IBM 与人工智能初创公司 Zyphra 共同发布全球首个全 AMD 硬件训练的 MoE 基础模型 ZAYA1。该模型预训练数据量达 14 万亿 token，综合性能与 Qwen3 系列持平，在数学/STEM 推理任务中无需指令微调即可接近 Qwen3 专业版水平。该模型基于 IBM Cloud 提供的 128 个计算节点（总计 1024 张 AMD ...</p>\n\n<a href=\"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN20251126060147977f4d30&s=b\">网页链接</a>\n\n</div>\n","source":"tencent","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>AMD 联合 IBM 与 Zyphra 发布全球首款全 AMD 硬件训练 MoE 模型 ZAYA1</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\">\nAMD 联合 IBM 与 Zyphra 发布全球首款全 AMD 硬件训练 MoE 模型 ZAYA1\n</h2>\n\n<h4 class=\"meta\">\n\n\n2025-11-26 05:55 北京时间&nbsp;&nbsp;&nbsp;<a href=http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN20251126060147977f4d30&s=b><strong>牛透社</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>牛透社 11 月 25 日消息，AMD 联合 IBM 与人工智能初创公司 Zyphra 共同发布全球首个全 AMD 硬件训练的 MoE 基础模型 ZAYA1。该模型预训练数据量达 14 万亿 token，综合性能与 Qwen3 系列持平，在数学/STEM 推理任务中无需指令微调即可接近 Qwen3 专业版水平。该模型基于 IBM Cloud 提供的 128 个计算节点（总计 1024 张 AMD ...</p>\n\n<a href=\"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN20251126060147977f4d30&s=b\">网页链接</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","isBrief":false,"type":0,"news_type":1,"symbol":"LU0889565833.HKD","symbol_name":"FRANKLIN TECHNOLOGY \"A\" (HKD) ACC","start_time":0,"source_url":"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN20251126060147977f4d30&s=b","article_id":"2586404116","we_media_id":null,"thumbnails":[],"rights":{"source":"tencent","url":"http://gu.qq.com/resources/shy/news/detail-v2/index.html#/?id=nesSN20251126060147977f4d30&s=b","rn_cache_url":null,"customStyle":"body{padding-top:10px;}#news_title{font-weight:bold;#titleStyle#;}#news_description span{font-size:12px;#descriptionStyle#;}.footer-note{#statement#}","selectors":".mod-LoadTzbdNews, body","filters":".relate-stock, .hot-list, .recom-box, .wx-sou","directOrigin":true},"url":"https://stock-news.laohu8.com/highlight/detail?id=2586404116","pubTimestamp":1764107700,"columns":[],"sourceInfo":{"source_id":"tencent","name":"腾讯"},"weMediaInfo":null,"summary":"牛透社 11 月 25 日消息，AMD 联合 IBM 与人工智能初创公司 Zyphra 共同发布全球首个全 AMD 硬件训练的 MoE 基础模型 ZAYA1。该模型基于 IBM Cloud 提供的 128 个计算节点，峰值算力达 750 千万亿次浮点运算，采用卷积压缩注意力机制使显存占用降低 32%，线性路由架构在 70% 稀疏度下实现路由精度提升 2.3 个百分点。AMD 表示已验证 MI300X 加速卡与 ROCm 生态的大规模训练能力，计划 2026 年实现百亿参数级 MoE 模型的总体拥有成本与 NVIDIA 方案持平。","collect":0,"end_time":0,"defaultTopTitle":"qq.com","property":[],"viewcount":null,"language":"zh","relate_stocks":{"LU0889565833.HKD":"FRANKLIN TECHNOLOGY \"A\" (HKD) ACC","BK4141":"半导体产品","LU0823421333.USD":"BNP PARIBAS DISRUPTIVE TECHNOLOGY \"C\" (USD) ACC","BK4534":"瑞士信贷持仓","BK4533":"AQR资本管理(全球第二大对冲基金)","LU0823421416.USD":"BNP PARIBAS DISRUPTIVE TECHNOLOGY \"C\" (USD) INC","BK4602":"量子计算概念","BK4529":"IDC概念","BK4566":"资本集团","LU1852331112.SGD":"Blackrock World Technology Fund A2 SGD-H","LU2458330169.SGD":"FRANKLIN SHARIAH TECHNOLOGY \"A\" (SGD) ACC","LU0321505868.SGD":"Schroder ISF Global Dividend Maximiser A Dis SGD","LU0868494617.USD":"UBS (LUX) EQUITY SICAV - US TOTAL YIELD SUSTAINABLE \"P\" (USD) ACC","LU1880398471.USD":"AMUNDI FUNDS GLOBAL EQUITY \"A2\" (USD) ACC","GFS":"GLOBALFOUNDRIES Inc.","LU1880398554.USD":"AMUNDI FUNDS GLOBAL EQUITY \"A2\" (USD) INC","LU0823434583.USD":"BNP PARIBAS US GROWTH \"C\" (USD) ACC","LU2458330243.SGD":"FRANKLIN SHARIAH TECHNOLOGY \"A-H1\" (SGDHDG) ACC","LU0109392836.USD":"富兰克林科技股A","IE0004445239.USD":"JANUS HENDERSON US FORTY \"A2\" (USD) ACC","LU1861558580.USD":"日兴方舟颠覆性创新基金B","LU1951198990.SGD":"Natixis Thematics AI & Robotics Fund H-R/A SGD-H","IBM":"IBM","BK4538":"云计算","LU0823434740.USD":"BNP PARIBAS US GROWTH \"C\" (USD) INC","AMD":"美国超微公司","LU1951200564.SGD":"Natixis Thematics AI & Robotics Fund R/A SGD","LU1242518931.SGD":"Fullerton Lux Funds - Asia Absolute Alpha A Acc SGD","LU0127658192.USD":"EASTSPRING INVESTMENTS GLOBAL TECHNOLOGY \"A\" (USD) ACC","LU2355687059.USD":"NORDEA 1 GLOBAL IMPACT  \"BP\" (USD) ACC","BK4605":"半导体精选","LU1894683348.USD":"AMUNDI FUNDS US EQUITY RESEARCH VALUE \"A2\" (USD) INC","LU0979878070.USD":"FULLERTON LUX FUNDS - ASIA ABSOLUTE ALPHA \"A\" (USD) ACC","BK4134":"信息科技咨询与其它服务","LU1989764664.SGD":"CPR Invest - Global Disruptive Opportunities A2 Acc SGD-H","LU1923623000.USD":"Natixis Thematics AI & Robotics Fund R/A USD","IE00BMPRXR70.SGD":"Neuberger Berman 5G Connectivity A Acc SGD-H","LU2098885051.SGD":"JPMorgan Funds - Multi-Manager Alternatives A (acc) SGD","BK4573":"虚拟现实","LU0081259029.USD":"UBS (LUX) EQUITY FUND - TECH OPPORTUNITY \"P\" (USD) ACC","IE00BMPRXN33.USD":"NEUBERGER BERMAN 5G CONNECTIVITY \"A\" (USD) ACC","LU0198837287.USD":"UBS (LUX) EQUITY SICAV - USA GROWTH \"P\" (USD) ACC","LU1303367103.USD":"摩根大通多经理另类基金 A (acc)","LU1674673428.USD":"HSBC GIF GLOBAL LOWER CARBON EQUITY \"AC\" (USD) ACC","BK4532":"文艺复兴科技持仓","LU1861559042.SGD":"日兴方舟颠覆性创新基金B SGD","LU1883839398.USD":"AMUNDI FUNDS INCOME OPPORTUNITIES \"A2\" (USD) ACC","BK4543":"AI","BK4512":"苹果概念","LU2264538146.SGD":"Fullerton Lux Funds - Global Absolute Alpha A Acc SGD"},"translate_title":"AMD, IBM and Zyphra release the world's first all-AMD hardware training MoE model ZAYA1","themeId":null,"isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{"IBM":1.5,"AMD":1.5,"GFS":1.5},"content_text":"牛透社 11 月 25 日消息，AMD 联合 IBM 与人工智能初创公司 Zyphra 共同发布全球首个全 AMD 硬件训练的 MoE 基础模型 ZAYA1。该模型预训练数据量达 14 万亿 token，综合性能与 Qwen3 系列持平，在数学/STEM 推理任务中无需指令微调即可接近 Qwen3 专业版水平。该模型基于 IBM Cloud 提供的 128 个计算节点（总计 1024 张 AMD Instinct MI300X 加速卡），峰值算力达 750 千万亿次浮点运算，采用卷积压缩注意力机制使显存占用降低 32%，线性路由架构在 70% 稀疏度下实现路由精度提升 2.3 个百分点。AMD 表示已验证 MI300X 加速卡与 ROCm 生态的大规模训练能力，计划 2026 年实现百亿参数级 MoE 模型的总体拥有成本与 NVIDIA 方案持平。","kind":"news","is_publish_news":true,"is_publish_highlight":false,"is_publish_live":false,"is_publish_wemedia":null,"editions":null,"column":"","sentiment":"1","news_tag":"productRelease","news_rank":0,"symbols":[],"gpt_button":0,"need_auth":false,"code":"91000000","status":"200"},"commentList":[],"isCommentEnd":true,"newsSizeData":{"likeSize":0,"commentSize":0,"repostSize":0,"favoriteSize":0,"likeStatus":false,"favoriteStatus":false},"APP":{"userAgent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)","isDev":false,"isTTM":false,"deviceId":"web-server-community-laohu8-v3","version":"4.35.9","shortVersion":"4.35.9","platform":"web","vendor":"web","appName":"laohu8","isIOS":false,"isAndroid":false,"isTiger":false,"isTHS":false,"isWeiXin":false,"isWeiXinMini":false,"isWeiBo":false,"isQQ":false,"isBaiduSwan":false,"isBaiduBox":false,"isDingTalk":false,"isToutiao":false,"isOnePlus":false,"isHuaWei":false,"isXiaomi":false,"isXiaomiWebView":false,"isOppo":false,"isVivo":false,"isSamsung":false,"isMobile":false},"href":"/m/news/2586404116","isCrawlerRequest":true}