{"id":404150,"date":"2026-08-19T09:32:32","date_gmt":"2026-08-19T08:32:32","guid":{"rendered":"https:\/\/droix.net\/blogs\/?p=365004"},"modified":"2026-09-28T15:41:42","modified_gmt":"2026-09-28T14:41:42","slug":"geekom-pristato-keturiu-mazgu-mini-kompiuteriu-ai-klasteri-su-vietiniu-deepseek-modeliu","status":"publish","type":"post","link":"https:\/\/droix.net\/blogs\/lt\/geekom-pristato-keturiu-mazgu-mini-kompiuteriu-ai-klasteri-su-vietiniu-deepseek-modeliu\/","title":{"rendered":"GEEKOM pristato keturi\u0173 mazg\u0173 mini kompiuteri\u0173 AI klaster\u012f su vietiniu DeepSeek modeliu"},"content":{"rendered":"\n

GEEKOM<\/a> pristat\u0117 keturi\u0173 mazg\u0173 AI klaster\u012f, sukurt\u0105 i\u0161 kompakti\u0161k\u0173 A9 Mega Mini kompiuteri\u0173, parodydama, kaip ma\u017eo formato sistemas galima sujungti vietin\u0117ms \u012fmoni\u0173 AI programoms.<\/p>\n\n\n\n

Klasteris jungia keturis A9 Mega<\/a> \u012frenginius per USB4, leisdamas jiems veikti kaip paskirstytai platformai nereikalaujant tradicin\u0117s serveri\u0173 spintos ar patentuotos didel\u0117s spartos tinklo \u012frangos. Kiekviena sistema varoma AMD Ryzen AI Max+ 395 <\/a>procesoriaus su 16 Zen 5 CPU branduoli\u0173, Radeon 8060S grafika ir vieninga atmintimi.<\/p>\n\n\n\n

\n
\"Keturi<\/a>
Keturi A9 Mega \u012frenginiai<\/figcaption><\/figure>\n\n\n\n
\"Keturi<\/a>
Keturi A9 Mega \u012frenginiai<\/figcaption><\/figure>\n<\/figure>\n\n\n\n

Naudojant Ubuntu, ROCm ir DwarfStar programin\u0119 \u012frang\u0105, klasteris paskirsto DeepSeek V4 Flash model\u012f visose keturiose ma\u0161inose. Jis taip pat teikia OpenAI suderinam\u0105 API, leid\u017eian\u010di\u0105 integruoti su programomis, AI agentais ir automatizavimo \u012frankiais.<\/p>\n\n\n\n

Pagrindinis \u0161ios s\u0105rankos privalumas yra vietinis apdorojimas. \u012emon\u0117s gali analizuoti dokumentus, \u0161altinio kod\u0105, sutartis ir vidines \u017einias nesi\u0173sdamos jautrios informacijos \u012f vie\u0161\u0105 debes\u012f. GEEKOM teigia, kad sistema taip pat buvo testuota su konteksto langais iki 250 000 \u017eeton\u0173, tod\u0117l ji tinka ilgiems dokumentams ir sud\u0117tingiems darbo procesams.<\/p>\n\n\n\n

\n
\"Keturi<\/a>
Keturi GEEKOM A9 Mega \u012frenginiai<\/figcaption><\/figure>\n\n\n\n
\"Keturi<\/a>
Keturi A9 Mega \u012frenginiai, vaizdas i\u0161 galo<\/figcaption><\/figure>\n<\/figure>\n\n\n\n

Pasak \u012fmon\u0117s, klasteris pasiek\u0117 ma\u017edaug 14,61 \u017eetono per sekund\u0119 esant vienkartiniam vienalaiki\u0161kumui, o 95-ojo procentilio laikas iki pirmojo \u017eetono buvo apie 0,42 sekund\u0117s atliekant 32 ir 128 \u017eeton\u0173 testus. Organizacijos gali prad\u0117ti nuo vienos ar dviej\u0173 A9 Mega<\/a> sistem\u0173 ir i\u0161pl\u0117sti iki keturi\u0173, kai auga j\u0173 poreikiai.<\/p>\n\n\n\n

Demonstracija i\u0161ry\u0161kina kompakti\u0161k\u0173 kompiuteri\u0173 potencial\u0105 kaip ple\u010diamas vietines AI platformas \u012fmon\u0117ms, laboratorijoms, klas\u0117ms ir kra\u0161tinio kompiuterijos aplinkoms.<\/p>\n\n\n\n

\u0160altinis: GEEKOM per PR Newswire<\/a>. Na\u0161umo rodikliai ir galimyb\u0117s pagr\u012fsti \u012fmon\u0117s pateiktais testavimo duomenimis.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

GEEKOM pristat\u0117 keturi\u0173 mazg\u0173 AI klaster\u012f, sukurt\u0105 i\u0161 kompakti\u0161k\u0173 A9 Mega Mini kompiuteri\u0173. Sujungtas per USB4 ir varomas AMD Ryzen AI Max+ 395 procesori\u0173, sistema vietoje paleid\u017eia DeepSeek V4 Flash, si\u016blydama \u012fmon\u0117ms ple\u010diam\u0105 b\u016bd\u0105 apdoroti jautrius dokumentus, kod\u0105 ir AI darbo kr\u016bvius nepasikliaujant vien tik debesijos infrastrukt\u016bra.<\/p>\n","protected":false},"author":859,"featured_media":404151,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2500,2985],"tags":[32171,3280,2615],"class_list":["post-404150","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-naujienos","category-mini-kompiuteris-lt","tag-a9-mega-15","tag-geekom-lt","tag-mini-kompiuteris"],"lang":"lt","translations":{"lt":404150,"en":365004,"de":370301,"en-gb":370672,"en-ie":370679,"en-au":370686,"fr":371387,"ar":382327,"bg":384272,"cs":385446,"es":387053,"da":387374,"pt-pt":388137,"hr":389905,"hu":390629,"lv":393581,"el":403888,"et":403954,"fi":404020,"it":404085,"nl":404259,"no":404325,"pl":404390,"ro":404456,"sk":404522,"sl":404588,"sv":404653},"pll_sync_post":{},"_links":{"self":[{"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/posts\/404150","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/users\/859"}],"replies":[{"embeddable":true,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/comments?post=404150"}],"version-history":[{"count":1,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/posts\/404150\/revisions"}],"predecessor-version":[{"id":404156,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/posts\/404150\/revisions\/404156"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/media\/404151"}],"wp:attachment":[{"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/media?parent=404150"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/categories?post=404150"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/droix.net\/blogs\/wp-json\/wp\/v2\/tags?post=404150"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}