{"id":6072,"date":"2025-01-20T05:49:00","date_gmt":"2025-01-19T20:49:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6072"},"modified":"2025-01-20T05:49:00","modified_gmt":"2025-01-19T20:49:00","slug":"minimax-01-scaling-foundation-models-with-lightning-attention","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6072","title":{"rendered":"MiniMax-01: Scaling Foundation Models with Lightning Attention\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>MiniMax-01: Scaling Foundation Models with Lightning Attention\u00a0<\/strong>[59.4]<br>MiniMax-Text-01\u3068MiniMax-VL-01\u306f\u3001\u3088\u308a\u9577\u3044\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u3092\u51e6\u7406\u3059\u308b\u306e\u306b\u512a\u308c\u305f\u6a5f\u80fd\u3092\u63d0\u4f9b\u3059\u308b\u3002 MiniMax-Text-01\u306f\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u4e2d\u306b\u6700\u5927100\u4e07\u306e\u30c8\u30fc\u30af\u30f3\u306b\u5230\u9054\u3067\u304d\u3001\u63a8\u8ad6\u6642\u306b400\u4e07\u306e\u30c8\u30fc\u30af\u30f3\u3092\u5b89\u4fa1\u306a\u4fa1\u683c\u3067\u5916\u633f\u3067\u304d\u308b\u3002 \u79c1\u305f\u3061\u306e\u30d3\u30b8\u30e7\u30f3\u8a00\u8a9e\u30e2\u30c7\u30eb\u3067\u3042\u308bMiniMax-VL-01\u306f\u3001512\u5104\u306e\u30d3\u30b8\u30e7\u30f3\u8a00\u8a9e\u30c8\u30fc\u30af\u30f3\u306b\u3088\u308b\u7d99\u7d9a\u7684\u306a\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306b\u3088\u3063\u3066\u69cb\u7bc9\u3055\u308c\u307e\u3059\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2501.08313v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2501.08313v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Tue, 14 Jan 2025 18:50:05 GMT)<\/li>\n\n\n\n<li>456B\uff0832\u30a8\u30ad\u30b9\u30d1\u30fc\u30c8\u3001\u30a2\u30af\u30c6\u30a3\u30d6\u30d1\u30e9\u30e1\u30fc\u30bf 45.9B\uff09\u306eMoE\u69cb\u6210\u306e\u5927\u898f\u6a21\u306a\u516c\u958bLLM\u3002\u6027\u80fd\u306fGPT-4o\u306a\u3069\u5546\u7528\u30e2\u30c7\u30eb\u306b\u5339\u6575\u3059\u308b\u3046\u3048\u3001\u6271\u3048\u308b\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u9577\u304c4M\u30c8\u30fc\u30af\u30f3\u3068\u3068\u3066\u3082\u9577\u3044\u3002\u300cWe demonstrate the first successful large-scale implementation of linear attention.\u300d\u3068\u4e3b\u5f35\uff08\u300cAfter extensive experimentation, we settled on a hybrid architecture mainly using lightning attention (Qin et al , 2024b), an I\/O-aware implementation of a linear attention variant (Qin et al , 2022a).\u300d\u3068\u3082\u3042\u308b\u901a\u308a\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u69cb\u6210\uff09\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/MiniMax-AI\/MiniMax-01\">GitHub &#8211; MiniMax-AI\/MiniMax-01<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[223,667],"class_list":["post-6072","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-667"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6072","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=6072"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6072\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6072"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6072"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6072"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}