{"id":5233,"date":"2024-03-11T09:33:53","date_gmt":"2024-03-11T01:33:53","guid":{"rendered":"https:\/\/www.1ai.net\/?p=5233"},"modified":"2024-03-11T09:33:53","modified_gmt":"2024-03-11T01:33:53","slug":"amd-ai%e8%8a%af%e7%89%87-instinct-mi300x-%e5%90%b8%e5%bc%95%e5%a4%a7%e9%87%8f%e7%94%a8%e6%88%b7%e5%bc%83%e7%94%a8%e8%8b%b1%e4%bc%9f%e8%be%be","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/5233.html","title":{"rendered":"AMD AI chip Instinct MI300X attracts a large number of users to abandon Nvidia"},"content":{"rendered":"<p data-vmark=\"696f\">According to a recent survey, a large number of AI professionals are planning to<a href=\"https:\/\/www.1ai.net\/en\/tag\/%e8%8b%b1%e4%bc%9f%e8%be%be\" title=\"Look at the article with the label\" target=\"_blank\" >Nvidia<\/a>Turn to adoption <a href=\"https:\/\/www.1ai.net\/en\/tag\/amd\" title=\"[SEE ARTICLES WITH [AMD] LABELS]\" target=\"_blank\" >AMD<\/a> Instinct MI300X GPU.<\/p>\n<p data-vmark=\"bab3\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-full wp-image-5234\" title=\"6796bf5d-b40e-40db-abb2-400a2e09a9e9\" src=\"https:\/\/www.1ai.net\/wp-content\/uploads\/2024\/03\/6796bf5d-b40e-40db-abb2-400a2e09a9e9.png\" alt=\"6796bf5d-b40e-40db-abb2-400a2e09a9e9\" width=\"728\" height=\"420\" \/><\/p>\n<p data-vmark=\"c9ca\">TensorWave co-founder Jeff Tatarchuk revealed that they conducted an independent survey covering 82 engineers and artificial intelligence professionals. Among them, about 50% of respondents said they were confident in the AMD Instinct MI300X GPU, because compared with Nvidia&#039;s H100 series products, the MI300X not only has a better price-performance ratio, but also has sufficient supply, avoiding the problem of tight supply. Jeff also mentioned that TensorWave will also adopt the MI300X artificial intelligence accelerator. This is undoubtedly good news for AMD, because in the past, its Instinct series products have always been at a disadvantage in market share compared to Nvidia&#039;s competitors.<\/p>\n<p data-vmark=\"3dd4\">The AMD Instinct MI300X AI accelerator is built on the CDNA 3 architecture, using a hybrid 5nm and 6nm process technology, with up to 153 billion transistors (MI300X model). Storage has also been greatly improved, with the MI300X&#039;s HBM3 capacity 50% higher than the previous generation MI250X (128 GB).<\/p>\n<p data-vmark=\"9dc3\">Compared with NVIDIA H100, MI300X has the following advantages:<\/p>\n<ul class=\"list-paddingleft-2\">\n<li>\n<p data-vmark=\"d22e\">2.4 times more memory capacity<\/p>\n<\/li>\n<li>\n<p data-vmark=\"9dc2\">1.6 times higher memory bandwidth<\/p>\n<\/li>\n<li>\n<p data-vmark=\"c382\">1.3 times higher FP8 performance (TFLOPS)<\/p>\n<\/li>\n<li>\n<p data-vmark=\"b2c9\">1.3x higher FP16 performance (TFLOPS)<\/p>\n<\/li>\n<li>\n<p data-vmark=\"159b\">In 1v1 comparison test, the performance can lead H100 (Llama 2 70B) by up to 20%<\/p>\n<\/li>\n<li>\n<p data-vmark=\"bf79\">In 1v1 comparison test, the performance can lead H100 (FlashAttention 2) by up to 20%<\/p>\n<\/li>\n<li>\n<p data-vmark=\"c860\">In the 8v8 server comparison test, the performance can lead H100 (Llama 2 70B) by up to 40%<\/p>\n<\/li>\n<li>\n<p data-vmark=\"1e74\">In the 8v8 server comparison test, the performance can lead H100 (Bloom 176B) by up to 60%<\/p>\n<\/li>\n<\/ul>\n<p data-vmark=\"b89e\">AMD&#039;s flagship Instinct AI accelerator, the MI300X, not only represents a huge leap in performance, but also comes at a very clever time, as Nvidia is currently plagued by a backlog of orders, which limits their opportunities to expand into new customers. Although AMD&#039;s performance in the field of artificial intelligence has not been satisfactory before, the future seems bright and it is expected to gain a foothold in the fierce competition.<\/p>","protected":false},"excerpt":{"rendered":"<p>According to a recent survey, a large number of AI professionals are planning to switch from NVIDIA to AMD Instinct MI300X GPUs. TensorWave co-founder, Jeff Tatarchuk, revealed that they conducted an independent survey covering 82 engineers and AI professionals. About 50% of the respondents expressed their confidence in the AMD Instinct MI300X GPUs because of the better price\/performance ratio compared to NVIDIA's H100 series, as well as the availability of the MI300X to avoid supply constraints.<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[1099,333,239],"collection":[],"class_list":["post-5233","post","type-post","status-publish","format-standard","hentry","category-news","tag-ai","tag-amd","tag-239"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/5233","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=5233"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/5233\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=5233"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=5233"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=5233"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=5233"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}