[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"content-doc-dcio85118779":3},{"user":4,"document":8,"mainDocument":27,"columnUrl":29,"subscription":30,"footer":42,"text":80},{"isAuthenticated":5,"isAdmin":5,"displayName":6,"avatarUrl":6,"nid":6,"groupLevel":7},false,"",-10,{"id":9,"fullTitle":10,"subTitle":6,"url":11,"columnId":12,"columnName":13,"columnUrl":14,"summary":6,"contentHtml":15,"mainContentHtml":6,"posterUrl":16,"createDate":17,"displayDate":18,"displayDateSlash":19,"pageviews":20,"tags":21,"hidden":5,"isSubContent":5,"replyDocOrTargetId":6,"contentType":23,"videoId":6,"liveVideoUrl":6,"useContentVideo":5,"duration":24,"price":24,"priceText":25,"priceBadgeText":25,"priceBadgeClass":26,"freeForMinGroupLevel":24,"redirectUrl":6,"readyToStream":5},"dcio85118779","OpenAI自研Jalapeno芯片打败英伟达GB300","\u002Fdoc\u002Fdcio85118779","col18178739ee","美股资讯","\u002Fcol\u002Fcol18178739ee","\u003Cp>\n\u003C\u002Fp>\u003Cp>\u003Cspan>OpenAI\u003Cspan>今天表示，与博通\u003C\u002Fspan>(AVGO)\u003Cspan>合作开发了新款\u003C\u002Fspan>Jalapeno\u003Cspan>芯片在测试中功耗效率和响应延迟上超越英伟达\u003C\u002Fspan>GB300\u003Cspan>。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003Cspan>\u003Cspan>凸显出该公司自主研发\u003C\u002Fspan>AI\u003Cspan>处理器方面取得的进展。\u003C\u002Fspan>OpenAI\u003Cspan>芯片业务负责人理查德·何（\u003C\u002Fspan>Richard Ho\u003Cspan>）表示，在测试中，\u003C\u002Fspan>Jalapeno\u003Cspan>处理器在两个项目中领先：单位功耗所能处理的\u003C\u002Fspan>AI\u003Cspan>工作量，以及返回响应的速度。此次测试将该芯片与英伟达\u003C\u002Fspan>GB300\u003Cspan>进行对比，后者是此次测试所采用的公开基准测试系统中表现最领先产品。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003Cspan>OpenAI\u003Cspan>与博通\u003C\u002Fspan>(AVGO)\u003Cspan>合作开发了\u003C\u002Fspan>Jalapeno\u003Cspan>。博通为众多客户生产定制芯片。两家公司去年宣布达成合作，并于今年\u003C\u002Fspan>6\u003Cspan>月强调了该处理器的开发速度，称其在创纪录的时间内完成开发。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003Cspan>\u003Cspan>这也是\u003C\u002Fspan>OpenAI\u003Cspan>自研\u003C\u002Fspan>Jalapeño\u003Cspan>这类\u003C\u002Fspan>AI\u003Cspan>芯片值得关注的地方。如果它主要服务于\u003C\u002Fspan>OpenAI\u003Cspan>自己的模型，就不需要像\u003C\u002Fspan>NVIDIA\u003Cspan>一样做一套尽可能通用的\u003C\u002Fspan>GPU\u003Cspan>。它可以针对\u003C\u002Fspan>Transformer\u003Cspan>、\u003C\u002Fspan>Attention\u003Cspan>、矩阵计算、\u003C\u002Fspan>KV Cache\u003Cspan>以及特定数据类型进行高度定制，把自己不需要的功能直接砍掉，从而在特定\u003C\u002Fspan>AI workload\u003Cspan>上实现更高的性能功耗比和更低的延迟。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003Cspan>\u003Cspan>因此，如果\u003C\u002Fspan>Jalapeño\u003Cspan>真的能够在\u003C\u002Fspan>OpenAI\u003Cspan>自己的测试中超过\u003C\u002Fspan>GB300\u003Cspan>，更准确的理解并不是“\u003C\u002Fspan>OpenAI\u003Cspan>已经造出了一颗全面超过\u003C\u002Fspan>NVIDIA\u003Cspan>的\u003C\u002Fspan>GPU\u003Cspan>”，而是\u003C\u002Fspan>OpenAI\u003Cspan>正在利用\u003C\u002Fspan>ASIC\u003Cspan>的定制化优势，针对自己的\u003C\u002Fspan>AI\u003Cspan>工作负载打造专用计算平台。这种方式有可能在特定任务上比通用\u003C\u002Fspan>GPU\u003Cspan>更高效。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003Cspan>\u003Cspan>不过，真正最难复制的仍然可能不是芯片本身，而是\u003C\u002Fspan>NVIDIA\u003Cspan>长期建立的\u003C\u002Fspan>CUDA\u003Cspan>软件生态。要让一颗\u003C\u002Fspan>AI\u003Cspan>芯片大规模替代\u003C\u002Fspan>GPU\u003Cspan>，需要同时解决芯片、编译器、运行时、\u003C\u002Fspan>Kernel\u003Cspan>、框架适配以及开发者生态等问题。因此，\u003C\u002Fspan>AI\u003Cspan>芯片的竞争最终很可能不是单纯的“谁的芯片算力更高”，而是硬件、内存、网络、软件和模型协同优化的系统级竞争。\u003C\u002Fspan>\u003C\u002Fspan>\u003C\u002Fp>\n\u003Cp>\u003C\u002Fp>","https:\u002F\u002Fwww.tradesmax.com\u002Fimages\u002Fa_Stock\u002FN\u002FNVDA\u002FNVDA.jpg","2026-08-25T15:03:46","2026.08.25","2026\u002F08\u002F25",39272,[22],"NVDA","Article",0,"免费","success",{"id":9,"fullTitle":10,"subTitle":6,"url":11,"columnId":12,"columnName":13,"columnUrl":14,"summary":6,"contentHtml":15,"mainContentHtml":6,"posterUrl":16,"createDate":17,"displayDate":18,"displayDateSlash":19,"pageviews":20,"tags":28,"hidden":5,"isSubContent":5,"replyDocOrTargetId":6,"contentType":23,"videoId":6,"liveVideoUrl":6,"useContentVideo":5,"duration":24,"price":24,"priceText":25,"priceBadgeText":25,"priceBadgeClass":26,"freeForMinGroupLevel":24,"redirectUrl":6,"readyToStream":5},[22],"\u002Fcol\u002Fstocknews",{"visible":5,"marketingHtml":31,"services":32,"recentDocuments":41},"\u003Cfigure class=\"image\">\u003Ca href=\"https:\u002F\u002Fstockwe.com\u002Fdoc\u002Fdcio537efad5\" rel=\"noopener noreferrer\">\u003Cimg src=\"\u002Fimg\u002Fstockwewebfiles\u002Fweb-202408-stk\u002F1586109431mceclip0.jpg\">\u003C\u002Fa>\u003C\u002Ffigure>\u003Cdiv class=\"text-center\">\u003Ch2 class=\"card-title mx-auto\">\u003Cbr>\u003Ca rel=\"noopener noreferrer\" href=\"https:\u002F\u002Fstockwe.com\u002Fdoc\u002Fdcio537efad5\">案例介绍：英伟达深度研究报告\u003C\u002Fa>\u003C\u002Fh2>\u003C\u002Fdiv>",[33,37],{"productId":34,"serviceName":35,"priceText":36},"prod_PPxdDdK87QaiLv","月付","$12.95美元",{"productId":38,"serviceName":39,"priceText":40},"prod_PPxeMs3bix1da5","年付","$149.00美元",[],{"links":43,"images":71,"summaryHtml":76,"aboutTitle":77,"aboutHtml":78,"copyrightHtml":79},[44,47,50,53,56,59,62,65,68],{"label":45,"url":46},"深度报告","\u002Fcol\u002FdepthReport",{"label":48,"url":49},"VIP会员","\u002Fvip",{"label":51,"url":52},"期权推荐","\u002FOption",{"label":54,"url":55},"低价暴涨股","\u002FPenny",{"label":57,"url":58},"AI智能体","\u002FAiAgent",{"label":60,"url":61},"常见问题","https:\u002F\u002Fstockwe.com\u002FFAQ",{"label":63,"url":64},"美股课程","\u002Fcol\u002Fvideos",{"label":66,"url":67},"免责声明","\u002Fdisclaimer",{"label":69,"url":70},"联系我们","\u002FContactUs",[72,73,74,75],"\u002Fimg\u002Fstockwebsiteblob\u002Fweb-202509-stk\u002FUploaderzic2tuwsol2_2025_09_11_18_21_07.gif","\u002Fimg\u002Fstockwebsiteblob\u002Fweb-202509-stk\u002FUploadercakzdvydksw_2025_09_03_09_00_56.png","\u002Fimg\u002Fstockwebsiteblob\u002Fweb-202509-stk\u002FUploadergtjyagwvoyk_2025_09_14_08_32_05.png","\u002Fimg\u002Fstockwebsiteblob\u002Fweb-202509-stk\u002FUploader3u0tt4jhlqh_2025_09_23_22_30_48.png","邮箱: buy@TradesMax.com 美国电话 626-378-3637","公司介绍","\u003Cp class=\"MsoNormal\">美股大数据 \u003Ca href=\"https:\u002F\u002Fstockwe.com\" rel=\"noopener\">StockWe.com\u003C\u002Fa> 是一个美国领先的金融和美股信息大数据提供商，紧盯华尔街金融市场和行情，2008年成立于美国硅谷，创始人是前纽约证券交易所资深分析师Ken，联合多位摩根斯坦利分析师，谷歌 Meta工程师利用AI和大数据，配合十多年美股实战经验和业内量化交易模型，每天处理海量股票数据：挖掘潜力大牛股，捕捉期权异动大单，实时主力资金流向、机构持仓变化、川普突发新闻，美股买卖信号第一时间发到您手机APP。\u003C\u002Fp>","专业美股投资者都在这里",{"loading":81,"search":82,"searchPlaceholder":82,"hotContent":83,"draft":84,"noData":85,"searchNoData":86,"courseContent":87,"more":88,"buyNow":89,"subscribeNow":90,"encoding":91,"paidContent":92},"Loading...","搜索","热门内容","草稿","目前没有任何内容公布","当前检索内容没有数据","课程内容","更多","立即购买后观看","- 立即订阅 -","视频编码中...","付费内容"]