{"id":114737,"date":"2024-04-03T00:25:44","date_gmt":"2024-04-02T22:25:47","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/databricks-otkryl-bolshuyu-yazykovuyu-model-dbrx-operezhayushhuyu-v-testah-gpt-3-5"},"modified":"2024-04-03T00:25:44","modified_gmt":"2024-04-02T22:25:47","slug":"databricks-otkryl-bolshuyu-yazykovuyu-model-dbrx-operezhayushhuyu-v-testah-gpt-3-5","status":"publish","type":"post","link":"https:\/\/prohoster.info\/hr\/blog\/news\/databricks-otkryl-bolshuyu-yazykovuyu-model-dbrx-operezhayushhuyu-v-testah-gpt-3-5","title":{"rendered":"Databricks je razvio veliki jezi\u010dni model DBRX, koji nadma\u0161uje GPT-3.5 u testovima.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Tvrtka Databricks najavila je otvaranje velike jezi\u010dne modele DBRX, koja se mo\u017ee koristiti za izradu chatbota koji odgovaraju na pitanja na prirodnom jeziku, rje\u0161avaju predlo\u017eene matemati\u010dke zadatke, sposobne generirati sadr\u017eaj na zadanu temu i stvarati kod na raznim programskim jezicima. Model je razvila tvrtka Mosaic ML, koju je Databricks kupio za 1,3 milijarde dolara. Za obuku kori\u0161ten je klaster od 3072 GPU NVIDIA H100 Tensor Core. Za pokretanje gotovog modela preporu\u010duje se 320GB memorije. <\/p>\n

Tijekom obuke modela kori\u0161tena je arhitektura MoE (Mixture of Experts), koja omogu\u0107uje dobivanje preciznijih procjena stru\u010dnjaka, zajedno s kolekcijom tekstova i koda od 12 TB. Veli\u010dina konteksta koji model DBRX uzima u obzir iznosi 32 tisu\u0107e tokena (broj tokena koje model mo\u017ee obraditi i zapamtiti prilikom generiranja teksta). Za usporedbu, veli\u010dina konteksta u modelima Google Gemini i OpenAI GPT-4 tako\u0111er je 32 tisu\u0107e tokena, Google Gemma \u2014 8 tisu\u0107a, dok model GPT-4 Turbo podr\u017eava 128 tisu\u0107a. <\/p>\n

Model obuhva\u0107a 132 milijarde parametara i podijeljen je na 16 stru\u010dnih mre\u017ea, od kojih se mo\u017ee koristiti najvi\u0161e 4 prilikom obrade zahtjeva (ukupno ne vi\u0161e od 36 milijardi parametara za svaki token). Za usporedbu, model GPT-4 se procjenjuje da uklju\u010duje 1,76 trilijuna parametara, nedavno otkriveni model Grok (X\/Twitter) \u2014 314 milijardi, GPT-3.5 \u2014 175 milijardi, YaLM (Yandex) \u2014 100 milijardi, LLaMA (Meta) \u2014 65 milijardi, GigaChat (Sber) \u2014 29 milijardi, Gemma (Google) \u2014 7 milijardi. <\/p>\n

Model i povezane komponente distribuiraju se pod licencom Databricks Open Model License, koja omogu\u0107uje kori\u0161tenje, reprodukciju, kopiranje, izmjenu i stvaranje derivativnih proizvoda, ali uz neka ograni\u010denja. Na primjer, licenca zabranjuje kori\u0161tenje DBRX-a, izvedenih modela i bilo kakvih rezultata na temelju njih za pobolj\u0161anje drugih jezi\u010dnih modela, osim DBRX-a. Licenca tako\u0111er zabranjuje kori\u0161tenje modela u podru\u010djima koja kr\u0161e zakone i propise. Derivativni modeli moraju se distribuirati pod istom licencom. Pri kori\u0161tenju u proizvodima i uslugama koje koristi vi\u0161e od 700 milijuna korisnika mjese\u010dno, potrebno je dobiti posebno odobrenje. <\/p>\n

Prema rije\u010dima stvaratelja modela, DBRX po svojim karakteristikama i mogu\u0107nostima nadma\u0161uje modele GPT-3.5 kompanije OpenAI i Grok-1 od Twittera, te mo\u017ee konkurirati modelu Gemini 1.0 Pro u testiranju razumijevanja jezika, mogu\u0107nosti pisanja koda na programskim jezicima i rje\u0161avanja matemati\u010dkih zadataka. U nekim primjenama, poput generiranja SQL upita, DBRX se pribli\u017eava po u\u010dinkovitosti modelu GPT-4 Turbo, koji prednja\u010di na tr\u017ei\u0161tu. Osim toga, model se od konkurenata izdvaja vrlo brzom obradom i omogu\u0107ava gotovo trenutne odgovore. Konkretno, DBRX mo\u017ee generirati tekst brzinom do 150 tokena u sekundi po korisniku, \u0161to je otprilike dvostruko br\u017ee od modela LLaMA2-70B.

\"Databricks<\/center>
\"Databricks<\/center> <\/p>\n

Dodatno, treba istaknuti objavu tehni\u010dkog opisa otvorenog velikog jezi\u010dnog modela InternLM2, koji se distribuira pod licencom Apache 2.0, dostupan je u varijantama s 20, 7 i 1.8 milijardi parametara. Model se razvija u \u0161angajskoj laboratoriji umjetne inteligencije uz sudjelovanje nekoliko kineskih sveu\u010dili\u0161ta i zna\u010dajan je zbog podr\u0161ke do 200K tokena konteksta i podr\u0161ke ne samo engleskom, ve\u0107 i kineskom jeziku. U mnogim testovima model je blizu GPT-4.

\"Databricks<\/center>
\"Databricks<\/center>
\"Databricks<\/center> <\/p>\n

Osim toga, prijavljeno je o razvoju 84 novih jezgara za mno\u017eenje matrica za alat llamafile, koji se razvija u Mozilli i omogu\u0107uje stvaranje univerzalnih izvr\u0161nih datoteka za pokretanje velikih jezi\u010dnih modela strojnog u\u010denja (LLM). Promjene su zna\u010dajno ubrzale rad modela u llamafile-u prilikom izvr\u0161avanja na CPU-u. Na primjer, izvo\u0111enje modela pri kori\u0161tenju llamafile-a sada je br\u017ee od 30% do 500% u usporedbi s kori\u0161tenjem llama.cpp, ovisno o okru\u017eenju, a u odnosu na biblioteku MKL, matricne operacije koje stanu u L2 cache se u novoj implementaciji izvr\u0161avaju dvostruko br\u017ee.
\n
Izvor: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u041a\u043e\u043c\u043f\u0430\u043d\u0438\u044f Databricks \u043e\u0431\u044a\u044f\u0432\u0438\u043b\u0430 \u043e\u0431 \u043e\u0442\u043a\u0440\u044b\u0442\u0438\u0438 \u0431\u043e\u043b\u044c\u0448\u043e\u0439 \u044f\u0437\u044b\u043a\u043e\u0432\u043e\u0439 \u043c\u043e\u0434\u0435\u043b\u0438 DBRX, \u043a\u043e\u0442\u043e\u0440\u0430\u044f \u043c\u043e\u0436\u0435\u0442 \u043f\u0440\u0438\u043c\u0435\u043d\u044f\u0442\u044c\u0441\u044f \u0434\u043b\u044f \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u044f \u0447\u0430\u0442\u0431\u043e\u0442\u043e\u0432, \u043e\u0442\u0432\u0435\u0447\u0430\u044e\u0449\u0438\u0445 \u043d\u0430 \u0432\u043e\u043f\u0440\u043e\u0441\u044b \u043d\u0430 \u0435\u0441\u0442\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u044f\u0437\u044b\u043a\u0435, \u0440\u0435\u0448\u0430\u044e\u0449\u0438\u0445 \u043f\u0440\u0435\u0434\u043b\u043e\u0436\u0435\u043d\u043d\u044b\u0435 \u043c\u0430\u0442\u0435\u043c\u0430\u0442\u0438\u0447\u0435\u0441\u043a\u0438\u0435 \u0437\u0430\u0434\u0430\u0447\u0438, \u0441\u043f\u043e\u0441\u043e\u0431\u043d\u044b\u0445 \u0433\u0435\u043d\u0435\u0440\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u043a\u043e\u043d\u0442\u0435\u043d\u0442 \u043d\u0430 \u0437\u0430\u0434\u0430\u043d\u043d\u0443\u044e \u0442\u0435\u043c\u0443 \u0438 \u0441\u043e\u0437\u0434\u0430\u0432\u0430\u0442\u044c \u043a\u043e\u0434 \u043d\u0430 \u0440\u0430\u0437\u043b\u0438\u0447\u043d\u044b\u0445 \u044f\u0437\u044b\u043a\u0430\u0445 \u043f\u0440\u043e\u0433\u0440\u0430\u043c\u043c\u0438\u0440\u043e\u0432\u0430\u043d\u0438\u044f. \u041c\u043e\u0434\u0435\u043b\u044c \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043d\u0430 \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0435\u0439 Mosaic ML, \u043a\u043e\u0442\u043e\u0440\u0430\u044f \u0431\u044b\u043b\u0430 \u043a\u0443\u043f\u043b\u0435\u043d\u0430 Databricks \u0437\u0430 1.3 \u043c\u043b\u0440\u0434 \u0434\u043e\u043b\u043b\u0430\u0440\u043e\u0432. \u0414\u043b\u044f \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u043b\u0441\u044f \u043a\u043b\u0430\u0441\u0442\u0435\u0440 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":114738,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-114737","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Databricks je otvorio veliki jezi\u010dni model DBRX, koji nadma\u0161uje testove GPT-3.5 | ProHoster","description":"Tvrtka Databricks objavila je otvaranje velikog jezi\u010dnog modela DBRX, koji se mo\u017ee koristiti za izradu chatbota koji odgovaraju na pitanja na prirodnom jeziku i rje\u0161avaju predlo\u017eene matemati\u010dke zadatke.","canonical_url":"https:\/\/prohoster.info\/hr\/blog\/news\/databricks-otkryl-bolshuyu-yazykovuyu-model-dbrx-operezhayushhuyu-v-testah-gpt-3-5","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"hr_HR","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47Databricks \u043e\u0442\u043a\u0440\u044b\u043b \u0431\u043e\u043b\u044c\u0448\u0443\u044e \u044f\u0437\u044b\u043a\u043e\u0432\u0443\u044e \u043c\u043e\u0434\u0435\u043b\u044c DBRX, \u043e\u043f\u0435\u0440\u0435\u0436\u0430\u044e\u0449\u0443\u044e \u0432 \u0442\u0435\u0441\u0442\u0430\u0445 GPT-3.5 | ProHoster","og:description":"\u041a\u043e\u043c\u043f\u0430\u043d\u0438\u044f Databricks \u043e\u0431\u044a\u044f\u0432\u0438\u043b\u0430 \u043e\u0431 \u043e\u0442\u043a\u0440\u044b\u0442\u0438\u0438 \u0431\u043e\u043b\u044c\u0448\u043e\u0439 \u044f\u0437\u044b\u043a\u043e\u0432\u043e\u0439 \u043c\u043e\u0434\u0435\u043b\u0438 DBRX, \u043a\u043e\u0442\u043e\u0440\u0430\u044f \u043c\u043e\u0436\u0435\u0442 \u043f\u0440\u0438\u043c\u0435\u043d\u044f\u0442\u044c\u0441\u044f \u0434\u043b\u044f \u0441\u043e\u0437\u0434\u0430\u043d\u0438\u044f \u0447\u0430\u0442\u0431\u043e\u0442\u043e\u0432, \u043e\u0442\u0432\u0435\u0447\u0430\u044e\u0449\u0438\u0445 \u043d\u0430 \u0432\u043e\u043f\u0440\u043e\u0441\u044b \u043d\u0430 \u0435\u0441\u0442\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u043e\u043c \u044f\u0437\u044b\u043a\u0435, \u0440\u0435\u0448\u0430\u044e\u0449\u0438\u0445 \u043f\u0440\u0435\u0434\u043b\u043e\u0436\u0435\u043d\u043d\u044b\u0435 \u043c\u0430\u0442\u0435\u043c\u0430\u0442\u0438\u0447\u0435\u0441\u043a\u0438\u0435 \u0437\u0430\u0434\u0430\u0447\u0438.","og:url":"https:\/\/prohoster.info\/hr\/blog\/news\/databricks-otkryl-bolshuyu-yazykovuyu-model-dbrx-operezhayushhuyu-v-testah-gpt-3-5","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2024-04-02T22:25:47+00:00","article:modified_time":"2024-04-02T22:25:47+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"114737","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":"default","schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-22 21:17:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2026-01-22 21:17:19","updated":"2026-01-22 21:17:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts\/114737","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/comments?post=114737"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/posts\/114737\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/media\/114738"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/media?parent=114737"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/categories?post=114737"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/hr\/wp-json\/wp\/v2\/tags?post=114737"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}