[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"portal-settings:stajic:sr":3,"public-menus:all":38,"post:what-is-rag-the-simplest-explanation-of-how-it-works:sr":205,"related:post:what-is-rag-the-simplest-explanation-of-how-it-works:sr:1":1800},{"statusCode":4,"data":5,"message":37},200,{"tenantId":6,"lang":7,"defaultLang":8,"siteUrl":9,"contactEmail":10,"brandName":11,"logoUrl":12,"siteName":11,"siteDescription":13,"ogImage":10,"robotsIndex":14,"socialLinks":10,"reservedSlugs":10,"seoPolicy":15},"stajic","sr","de","https:\u002F\u002Fstajic.de",null,"Stajic Platform","\u002FLogo_Planet.svg","Stajic Portal",true,{"branding":16,"relatedContent":17,"crossDomainLinks":18},{"logoUrl":12},{"enabled":14},[19,22,25,28,31,34],{"url":20,"label":21,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Ffigure.rocks","figure.rocks",{"url":23,"label":24,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Floving.rocks","loving.rocks",{"url":26,"label":27,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.com","bazify.com",{"url":29,"label":30,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.de","bazify.de",{"url":32,"label":33,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.at","bazify.at",{"url":35,"label":36,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.ba","bazify.ba","Portal settings resolved",[39,45],{"id":40,"name":41,"location":42,"isActive":14,"isDefault":43,"items":44},1,"main-navigation","header",false,[],{"id":46,"name":47,"location":48,"isActive":14,"isDefault":14,"items":49},4,"main-menu","sidebar",[50,66,79,93,103,118,133],{"id":51,"title":52,"url":60,"target":61,"icon":62,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":64,"portfolioId":10,"children":65},"item-18",{"de":53,"en":54,"es":55,"fr":56,"it":54,"ru":57,"sr":58,"zh":59},"Startseite","Home","Inicio","Accueil","Главная","Почетна","首页","\u002Ffull-stack-web-developer-munich-performance-seo-and-maintainable-builds","_self","i-lucide-home","page",111,[],{"id":67,"title":68,"url":75,"target":61,"icon":76,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":77,"portfolioId":10,"children":78},"item-22",{"de":69,"en":69,"es":70,"fr":69,"it":71,"ru":72,"sr":73,"zh":74},"Vision","Visión","Visione","Видение","Визија","想象","\u002Fueber-uns-webdesign-muenchen-webaplikation","i-lucide-eye",113,[],{"id":80,"title":81,"url":89,"target":61,"icon":90,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":91,"portfolioId":10,"children":92},"item-19",{"de":82,"en":83,"es":84,"fr":83,"it":85,"ru":86,"sr":87,"zh":88},"Leistungen","Services","Servicios","Servizi","Услуги","Услуге","服务","\u002Fservices-dienstleistungen-muenchen","i-lucide-wrench",116,[],{"id":94,"title":95,"url":99,"target":61,"icon":100,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":101,"portfolioId":10,"children":102},"item-23",{"de":96,"en":96,"es":96,"fr":96,"it":96,"ru":97,"sr":97,"zh":98},"Blog","Блог","博客","\u002Fblog","i-lucide-book-open",112,[],{"id":104,"title":105,"url":114,"target":61,"icon":115,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":116,"portfolioId":10,"children":117},"item-32",{"de":106,"en":107,"es":108,"fr":109,"it":110,"ru":111,"sr":112,"zh":113},"Neue Technologien","New Technologies","Nuevas tecnologías","Nouvelles technologies","Nuove tecnologie","Новые технологии","Нове технологије","新技术！","\u002Fneue-webtechnologien","i-lucide-sparkles",122,[],{"id":119,"title":120,"url":129,"target":61,"icon":130,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":131,"portfolioId":10,"children":132},"item-20",{"de":121,"en":122,"es":123,"fr":124,"it":125,"ru":126,"sr":127,"zh":128},"Kontakt","Contact us!","Contacto","Contact","Contatto","Контакт","Контактирајте нас","联系我们！","\u002Fcontact","i-lucide-mail",115,[],{"id":134,"title":135,"url":144,"target":61,"icon":145,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":147},"item-21",{"de":136,"en":137,"es":138,"fr":139,"it":140,"ru":141,"sr":142,"zh":143},"Unsere Arbeit","Our Work","Nuestro trabajo","Nos réalisations","I nostri lavori","Наши работы","Наши радови","文件夹","\u002Fportfolio","i-lucide-briefcase",114,[148,161,175,181,193],{"id":149,"title":150,"url":144,"target":61,"icon":159,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":160},"item-24",{"de":151,"en":152,"es":153,"fr":154,"it":155,"ru":156,"sr":157,"zh":158},"Alle Projekte","All Projects","Todos los proyectos","Tous les projets","Tutti i progetti","Все проекты","Сви пројекти","所有项目","i-lucide-grid-3x3",[],{"id":162,"title":163,"url":171,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":174},"item-29",{"de":164,"en":165,"es":166,"fr":167,"it":168,"ru":169,"sr":170,"zh":143},"Local Roots, Global Reach","Local Roots - Global Reach","Empresa local ","Entreprise locale","Azienda locale","Местная компания","Локално предузеће глобално тржиште","\u002Fportfolio\u002Flocal-roots-global-reach-communication-media-systems-for-modern-business","i-lucide-folder","custom",[],{"id":176,"title":177,"url":179,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":180},"item-28",{"de":178,"en":178,"es":178,"fr":178,"it":178,"ru":178,"sr":178,"zh":178},"Solr Suggester","\u002Fportfolio\u002Fsolr-fuzzy-suggester-und-solr-infix-suggester-abfrage-ueber-ajax-und-filterung",[],{"id":182,"title":183,"url":191,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":192},"item-27",{"de":184,"en":185,"es":186,"fr":187,"it":188,"ru":189,"sr":190,"zh":185},"Firmenwebseite SEO","Company Website SEO","Sitio web corporativo SEO","Site web d’entreprise SEO","Sito web aziendale SEO","Корпоративный сайт SEO","Пословна веб-страница SEO","\u002Fportfolio\u002Fseo-sem-branding-mobile-webseite-muenchen",[],{"id":194,"title":195,"url":203,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":204},"item-31",{"de":196,"en":197,"es":198,"fr":199,"it":200,"ru":201,"sr":202,"zh":197},"Digitalisierungsportal","Digitalization Portal","Portal de digitalización","Portail de numérisation","Portale di digitalizzazione","Портал цифровизации","Портал за дигитализацију","\u002Fportfolio\u002Fdigitalisierungsportal-archiv-museum-bibliothek-ead-lido-mets-mods",[],{"statusCode":4,"data":206,"message":1799},{"id":207,"title":208,"slug":209,"content":210,"contentJson":211,"excerpt":874,"featuredImage":875,"featuredImageAlt":876,"featuredImageCaption":10,"featuredImageTitle":10,"featuredImageCopyright":10,"featuredImageAuthor":10,"featuredImageSourceUrl":10,"featuredImageLicense":10,"featuredImageIsAiGenerated":43,"status":877,"publishedAt":878,"createdAt":879,"updatedAt":880,"seoLocalePaths":881,"categories":890,"author":915,"translations":920},"478","Šta je RAG? Najjednostavnije objašnjenje kako funkcioniše","what-is-rag-the-simplest-explanation-of-how-it-works","\u003Cp>RAG zvuči komplikovano jer je ime komplikovano. Ideja nije. RAG jednostavno znači: pre nego što AI odgovori, prvo pronađe relevantne informacije iz izvora znanja i daje te informacije jezičkom modelu.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--info my-6 rounded-xl border p-5 border-blue-300 bg-blue-50 dark:border-blue-900 dark:bg-blue-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">RAG u jednoj rečenici\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>RAG je korak u kojem AI pretražuje bazu znanja za korisne informacije pre nego što LLM napiše odgovor.\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Cp>Zamislite LLM kao pametnu osobu koja sedi za stolom. RAG je bibliotekar koji donosi pravu stranicu iz prave knjige. LLM zatim čita tu stranicu i odgovara vam.\u003C\u002Fp>\n\u003Cnav class=\"editorjs-toc\" data-editorjs-toc=\"true\" aria-label=\"Sadržaj\">\u003Cstrong class=\"editorjs-toc__title\">Sadržaj\u003C\u002Fstrong>\u003Col class=\"editorjs-toc__list editorjs-toc__list--depth-0\">\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-5\" class=\"editorjs-toc__link\">Prvo: šta radi LLM?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-10\" class=\"editorjs-toc__link\">Zatim: šta je baza znanja?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-15\" class=\"editorjs-toc__link\">Dakle, šta RAG zapravo radi?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-19\" class=\"editorjs-toc__link\">Vrlo jednostavan primer\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-25\" class=\"editorjs-toc__link\">Sada važan deo: RAG nije trenutno stanje\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-30\" class=\"editorjs-toc__link\">Šta je baza podataka stanja?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-36\" class=\"editorjs-toc__link\">Kako tri dela rade zajedno\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-40\" class=\"editorjs-toc__link\">Da li RAG uvek koristi vektorsku bazu podataka?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-45\" class=\"editorjs-toc__link\">Šta je embedding, jednostavno rečeno?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-50\" class=\"editorjs-toc__link\">RAG takođe nije memorija\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-54\" class=\"editorjs-toc__link\">Pravi primer iz igre: PUBG Ally\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-60\" class=\"editorjs-toc__link\">Jedan kompletan primer\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-64\" class=\"editorjs-toc__link\">Zašto uopšte koristiti RAG?\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-68\" class=\"editorjs-toc__link\">Šta RAG ne garantuje\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-72\" class=\"editorjs-toc__link\">Najlakši mentalni model za pamćenje\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-75\" class=\"editorjs-toc__link\">Zaključak\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-79\" class=\"editorjs-toc__link\">Često postavljana pitanja\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-81\" class=\"editorjs-toc__link\">Pojmovnik\u003C\u002Fa>\u003C\u002Fli>\u003Cli class=\"editorjs-toc__item\">\u003Ca href=\"#section-83\" class=\"editorjs-toc__link\">Primarni izvori\u003C\u002Fa>\u003C\u002Fli>\u003C\u002Fol>\u003C\u002Fnav>\n\u003Ch2 id=\"section-5\">Prvo: šta radi LLM?\u003C\u002Fh2>\n\u003Cp>LLM je deo koji razume jezik i proizvodi jezik. Može da pročita vaše pitanje, razume instrukcije, uporedi informacije, objasni nešto i napiše odgovor.\u003C\u002Fp>\n\u003Cp>Ali LLM ne zna automatski šta se trenutno nalazi u bazi podataka vaše kompanije, vašoj igračkoj sesiji, vašim privatnim dokumentima ili datoteci koju ste kreirali pre pet minuta.\u003C\u002Fp>\n\u003Cp>Zna samo ono što je već unutar modela plus sve informacije koje mu aplikacija pruži u trenutnom zahtevu.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Jednostavno pravilo\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">LLM \u003Cstrong>misli i piše\u003C\u002Fstrong>. Ne poseduje automatski sve vaše trenutne podatke.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-10\">Zatim: šta je baza znanja?\u003C\u002Fh2>\n\u003Cp>Baza znanja je jednostavno informacija koju aplikacija može da pretražuje.\u003C\u002Fp>\n\u003Cp>Može da sadrži PDF-ove, priručnike, dokumentaciju proizvoda, članke podrške, ugovore, pravila igre, podatke o oružju, interne dokumente kompanije, zapise baze podataka ili drugi tekst.\u003C\u002Fp>\n\u003Cp>Baza znanja može biti lokalna na vašem računaru. Može biti na serveru. Može biti u vektorskoj bazi podataka. Takođe može biti izgrađena od običnih datoteka. RAG ne znači Internet.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--success my-6 rounded-xl border p-5 border-emerald-300 bg-emerald-50 dark:border-emerald-900 dark:bg-emerald-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Važno\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>RAG ne zahteva Internet.\u003C\u002Fstrong> Informacije mogu biti potpuno lokalne.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-15\">Dakle, šta RAG zapravo radi?\u003C\u002Fh2>\n\u003Csection class=\"editorjs-process my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Ceo RAG proces\u003C\u002Fh3>\u003Cdiv class=\"flex flex-col sm:flex-row gap-3\">\u003Cdiv class=\"editorjs-process__step min-w-0 flex-1 rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">1\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">1. Postavljate pitanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Na primer: Koju municiju koristi ovo oružje?\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__arrow shrink-0 self-center text-xl text-gray-400 rotate-90 sm:rotate-0\" aria-hidden=\"true\">→\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0 flex-1 rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">2\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">2. RAG pretražuje bazu znanja\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Sistem traži male delove informacija koji su najrelevantniji za vaše pitanje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__arrow shrink-0 self-center text-xl text-gray-400 rotate-90 sm:rotate-0\" aria-hidden=\"true\">→\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0 flex-1 rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">3\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">3. RAG daje te delove LLM-u\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">LLM prima pitanje plus pronađene informacije.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__arrow shrink-0 self-center text-xl text-gray-400 rotate-90 sm:rotate-0\" aria-hidden=\"true\">→\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0 flex-1 rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">4\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">4. LLM piše odgovor\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Koristi pronađene informacije kao kontekst za odgovor.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Cp>To je RAG.\u003C\u002Fp>\n\u003Cp>Puno ime je Retrieval-Augmented Generation. Retrieval znači pronalaženje relevantnih informacija. Augmented znači dodavanje tih informacija u kontekst modela. Generation znači da LLM piše konačni odgovor.\u003C\u002Fp>\n\u003Ch2 id=\"section-19\">Vrlo jednostavan primer\u003C\u002Fh2>\n\u003Cp>Zamislite da imate lokalnu bazu znanja o igri.\u003C\u002Fp>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Baza znanja sadrži\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Primer\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Oružja\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">AKM koristi municiju 7.62 mm\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Predmeti za lečenje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Med Kit obnavlja zdravlje\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Dodaci\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ovaj dodatak radi sa ovim oružjima\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Pravila mape\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Ova zona se ponaša na ovaj način\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Cp>Pitate: „Koju municiju koristi AKM?“\u003C\u002Fp>\n\u003Cp>RAG pretražuje bazu znanja i pronalazi unos o AKM. Taj mali deo informacije daje LLM-u. LLM zatim odgovara: „AKM koristi municiju 7.62 mm.“\u003C\u002Fp>\n\u003Cp>LLM nije bio potreban cela baza podataka. RAG je doneo samo koristan deo.\u003C\u002Fp>\n\u003Ch2 id=\"section-25\">Sada važan deo: RAG nije trenutno stanje\u003C\u002Fh2>\n\u003Cp>Ovde mnoga objašnjenja postaju zbunjujuća.\u003C\u002Fp>\n\u003Cp>RAG obično daje AI znanje. Sistem stanja daje AI činjenice o tome šta je istinito upravo sada.\u003C\u002Fp>\n\u003Csection class=\"editorjs-comparison my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Znanje naspram trenutnog stanja\u003C\u002Fh3>\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left dark:border-gray-700 dark:bg-gray-900\">\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">RAG \u002F znanje\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">Trenutno stanje\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Oružje\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Municija\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Zdravlje\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Neprijatelj\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Caside class=\"editorjs-callout editorjs-callout--warning my-6 rounded-xl border p-5 border-amber-300 bg-amber-50 dark:border-amber-900 dark:bg-amber-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Ne mešajte ovo dvoje\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">RAG odgovara: \u003Cstrong>Šta je generalno istinito?\u003C\u002Fstrong>\u003Cbr>Stanje odgovara: \u003Cstrong>Šta je istinito upravo sada?\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-30\">Šta je baza podataka stanja?\u003C\u002Fh2>\n\u003Cp>Baza podataka stanja ili skladište stanja je jednostavno mesto gde aplikacija čuva trenutne činjenice.\u003C\u002Fp>\n\u003Cp>U igri, engine već zna stvari kao što su vaše zdravlje, pozicija, inventar, municija, trenutna misija, obližnji objekti i status neprijatelja. AI sistem može izložiti odabrane delove tog stanja modelu.\u003C\u002Fp>\n\u003Cp>U poslovnoj aplikaciji, ista ideja može biti baza podataka narudžbina, zapis o kupcu, status projekta ili trenutna vrednost senzora.\u003C\u002Fp>\n\u003Cp>Stanje kreira sama aplikacija dok se stvari dešavaju. Ako izgubite zdravlje, igra ažurira vrednost zdravlja. Ako pokupite municiju, inventar se menja. Ako je narudžbina plaćena, poslovni sistem menja status narudžbine.\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--info my-6 rounded-xl border p-5 border-blue-300 bg-blue-50 dark:border-blue-900 dark:bg-blue-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Jednostavno pravilo\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">Aplikacija kreira i ažurira \u003Cstrong>stanje\u003C\u002Fstrong>. RAG pretražuje \u003Cstrong>znanje\u003C\u002Fstrong>. LLM koristi oboje da odluči šta da kaže ili uradi.\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-36\">Kako tri dela rade zajedno\u003C\u002Fh2>\n\u003Csection class=\"editorjs-process my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">LLM + stanje + RAG\u003C\u002Fh3>\u003Cdiv class=\"grid grid-cols-1 md:grid-cols-2 xl:grid-cols-3 gap-4\">\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">1\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">1. Trenutno stanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Aplikacija govori AI-ju šta je sada istina: zdravlje 41%, AKM opremljen, 23 metka.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">2\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">2. RAG\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Sistem pronalazi korisno znanje: kako oružje funkcioniše, koji predmet za lečenje je dostupan ili relevantno pravilo.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">3\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">3. LLM\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Model prima pitanje, trenutno stanje i pronađeno znanje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">4\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">4. Rezonovanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">LLM kombinuje te ulaze i odlučuje koji odgovor ili akcija na visokom nivou ima smisla.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">5\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">5. Aplikacija\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Ako je potrebna akcija, aplikacija ili game engine je izvršava i ponovo ažurira stanje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Cp>Dakle, osnovna arhitektura je:\u003C\u002Fp>\n\u003Caside class=\"editorjs-callout editorjs-callout--note my-6 rounded-xl border p-5 border-gray-300 bg-gray-50 dark:border-gray-700 dark:bg-gray-900\u002F40\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Najjednostavnija arhitektura\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>Stanje = šta je sada istina\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = korisno znanje\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>LLM = razume, rezonuje i piše\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Aplikacija = izvršava pravu akciju\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-40\">Da li RAG uvek koristi vektorsku bazu podataka?\u003C\u002Fh2>\n\u003Cp>Ne.\u003C\u002Fp>\n\u003Cp>Vektorska baza podataka je uobičajen način za izgradnju semantičke pretrage, ali to nije definicija RAG-a.\u003C\u002Fp>\n\u003Cp>Važan deo je pronalaženje: sistem pronalazi relevantne spoljne informacije i dodaje ih u kontekst LLM-a pre nego što se odgovor generiše.\u003C\u002Fp>\n\u003Cp>OpenAI-jev File Search, na primer, može da radi sa fajlovima sačuvanim u vektorskim skladištima. Fajlovi se dele na manje delove kako bi sistem mogao da pronađe delove koji su relevantni za pitanje. To je jedna implementacija iste osnovne ideje.\u003C\u002Fp>\n\u003Ch2 id=\"section-45\">Šta je embedding, jednostavno rečeno?\u003C\u002Fh2>\n\u003Cp>Ne morate da razumete embedding-e da biste razumeli RAG.\u003C\u002Fp>\n\u003Cp>Ali jednostavna verzija je ovo: embedding je numerička reprezentacija značenja. Pomaže sistemu za pretragu da pronađe tekst koji je konceptualno sličan čak i kada reči nisu potpuno iste.\u003C\u002Fp>\n\u003Cp>Na primer, obična pretraga po ključnim rečima može da traži tačne reči „popravka automobila“. Semantička pretraga takođe može da razume da se „popravi moje vozilo“ odnosi na sličnu temu.\u003C\u002Fp>\n\u003Cp>To čini embedding-e korisnim za RAG, ali RAG takođe može da koristi pretragu po ključnim rečima, upite baze podataka ili hibrid nekoliko metoda.\u003C\u002Fp>\n\u003Ch2 id=\"section-50\">RAG takođe nije memorija\u003C\u002Fh2>\n\u003Cp>Memorija je još jedan koncept koji se često meša sa RAG-om.\u003C\u002Fp>\n\u003Cp>Memorija je obično informacija koju sistem čuva o prethodnim interakcijama ili prethodnim događajima. RAG je mehanizam koji se koristi za pronalaženje relevantnog znanja kada je potrebno.\u003C\u002Fp>\n\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Deo\u003C\u002Fth>\u003Cth class=\"border border-gray-300 px-4 py-2 text-left font-semibold\">Jednostavno značenje\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">LLM\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Deo koji razume i generiše jezik\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">RAG\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Deo koji traži relevantno znanje pre odgovora\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Baza znanja\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Informacije koje RAG može da pretražuje\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Stanje\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Šta je trenutno istina u aplikaciji ili svetu\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Memorija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Informacije sačuvane iz prethodnih interakcija ili događaja\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Alat \u002F akcija\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Nešto što AI sme da pozove ili zatraži od aplikacije da uradi\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Kontekst\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-2\">Informacije trenutno stavljene pred LLM za ovaj zahtev\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"section-54\">Pravi primer iz igre: PUBG Ally\u003C\u002Fh2>\n\u003Cp>PUBG Ally je koristan primer jer čini razliku vidljivom.\u003C\u002Fp>\n\u003Cp>KRAFTON opisuje stanje meča uživo kao poseban izvor istine. Igra izlaže trenutne činjenice kroz alate za posmatranje: trenutno oružje, municiju, zdravlje, status sigurne zone, obližnje predmete i borbenu situaciju.\u003C\u002Fp>\n\u003Cp>Pretraga znanja je drugačiji posao. Sistem može da koristi pripremljeno znanje o oružju, dodacima, predmetima i pravilima. NVIDIA ACE Game Agent SDK takođe izlaže poseban RAG API za pronalaženje znanja iz baza koje su napravili programeri.\u003C\u002Fp>\n\u003Cp>To nam daje jasnu podelu: game engine kaže šta se sada dešava, pretraga obezbeđuje relevantno znanje, a jezički model odlučuje šta informacije znače.\u003C\u002Fp>\n\u003Caside class=\"editorjs-referral my-6\">\u003Ca href=\"https:\u002F\u002Ffigure.rocks\u002Fblog\u002Fpubg-ally-shows-why-ai-teammates-need-two-brains-fast-reflexes-and-slow-reasoning\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"flex flex-col sm:flex-row gap-4 rounded-xl border border-gray-200 dark:border-gray-700 p-4 transition hover:border-primary-500\">\u003Cdiv class=\"min-w-0 flex-1\">\u003Cstrong class=\"block text-lg text-gray-900 dark:text-gray-100\">PUBG Ally pokazuje zašto AI saigračima trebaju dva mozga: brzi refleksi i sporo rezonovanje\u003C\u002Fstrong>\u003Cp class=\"mt-2 text-sm text-gray-600 dark:text-gray-300\">Praktičan primer iz igre koji pokazuje kako stanje uživo, jezičko rezonovanje i deterministička kontrola na strani igre mogu da rade zajedno.\u003C\u002Fp>\u003Cspan class=\"mt-3 inline-flex text-sm font-medium text-primary-600 dark:text-primary-400\">Pročitajte članak o arhitekturi PUBG Ally →\u003C\u002Fspan>\u003C\u002Fdiv>\u003C\u002Fa>\u003C\u002Faside>\n\u003Ch2 id=\"section-60\">Jedan kompletan primer\u003C\u002Fh2>\n\u003Cp>Zamislite da kažete AI saigraču: „Nisko mi je zdravlje. Treba li da napadnemo?“\u003C\u002Fp>\n\u003Csection class=\"editorjs-process my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Šta se dešava dalje\u003C\u002Fh3>\u003Cdiv class=\"grid grid-cols-1 md:grid-cols-2 xl:grid-cols-3 gap-4\">\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">1\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">Stanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Igra prijavljuje: zdravlje 24%, jedan neprijatelj u blizini, dva predmeta za lečenje dostupna.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">2\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">RAG\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Sistem znanja pronalazi relevantna pravila za predmet za lečenje i možda informacije o trenutnom oružju ili taktičkom mehanizmu.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">3\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">LLM\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Model kombinuje vaš zahtev, trenutno stanje i pronađeno znanje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">4\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">Odluka\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Zaključuje da je lečenje prvo sigurnije od trenutnog napada.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">5\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">Alat \u002F game engine\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Agent zahteva legalnu akciju u igri kao što je premeštanje u zaklon ili korišćenje predmeta za lečenje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv class=\"editorjs-process__step min-w-0  rounded-xl border border-gray-200 dark:border-gray-700 p-4\">\u003Cdiv class=\"text-xs font-semibold text-gray-500 dark:text-gray-400\">6\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 font-semibold text-gray-900 dark:text-gray-100\">Novo stanje\u003C\u002Fdiv>\u003Cdiv class=\"mt-1 text-sm text-gray-600 dark:text-gray-300\">Igra izvršava akciju i prijavljuje ažuriranu situaciju nazad agentu.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Cp>RAG nije kontrolisao lika. Baza podataka o stanju nije rezonovala. LLM nije direktno menjao igru. Svaki deo je imao jedan zadatak.\u003C\u002Fp>\n\u003Ch2 id=\"section-64\">Zašto uopšte koristiti RAG?\u003C\u002Fh2>\n\u003Cp>Zato što bi stavljanje svakog dokumenta, pravila i zapisa iz baze u svaki prompt bilo sporo, skupo i često zbunjujuće.\u003C\u002Fp>\n\u003Cp>RAG omogućava sistemu da izabere samo informacije koje su korisne za trenutno pitanje.\u003C\u002Fp>\n\u003Cp>Takođe vam omogućava da ažurirate bazu znanja bez ponovnog treniranja celog jezičkog modela. Promenite dokument ili bazu podataka, ponovo izgradite ili osvežite indeks kada je potrebno, i sledeća pretraga može da koristi novije informacije.\u003C\u002Fp>\n\u003Ch2 id=\"section-68\">Šta RAG ne garantuje\u003C\u002Fh2>\n\u003Cp>RAG može da poboljša utemeljenje, ali ne čini odgovor automatski tačnim.\u003C\u002Fp>\n\u003Cp>Korak pretrage može da pronađe pogrešan dokument. Tačan dokument može biti zastareo. LLM može da pogrešno razume dobre dokaze. Ili se trenutno stanje može promeniti.\u003C\u002Fp>\n\u003Cp>Pouzdan sistem stoga mora odvojeno da validira pretragu, svežinu stanja i konačno rezonovanje modela.\u003C\u002Fp>\n\u003Ch2 id=\"section-72\">Najlakši mentalni model za pamćenje\u003C\u002Fh2>\n\u003Csection class=\"editorjs-comparison my-6\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Zamislite AI sistem kao osobu za radnim stolom\u003C\u002Fh3>\u003Cdiv class=\"overflow-x-auto\">\u003Ctable class=\"w-full border-collapse\">\u003Cthead>\u003Ctr>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left dark:border-gray-700 dark:bg-gray-900\">\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">Analogija\u003C\u002Fth>\u003Cth class=\"border border-gray-300 bg-gray-50 px-4 py-3 text-left font-semibold dark:border-gray-700 dark:bg-gray-900\">AI sistem\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Osoba koja razmišlja\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Pronalaženje priručnika\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Knjige na polici\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Trenutna kontrolna tabla ili instrument tabla\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Beleške sa ranijih sastanaka\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Cth class=\"border border-gray-300 px-4 py-3 text-left font-semibold dark:border-gray-700\">Radi nešto u stvarnom svetu\u003C\u002Fth>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003Ctd class=\"border border-gray-300 px-4 py-3 dark:border-gray-700\">\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Caside class=\"editorjs-callout editorjs-callout--success my-6 rounded-xl border p-5 border-emerald-300 bg-emerald-50 dark:border-emerald-900 dark:bg-emerald-950\u002F20\" role=\"note\">\u003Cstrong class=\"block mb-2 text-gray-900 dark:text-gray-100\">Ako zapamtite samo ovo\u003C\u002Fstrong>\u003Cdiv class=\"text-gray-700 dark:text-gray-200\">\u003Cstrong>LLM = mozak.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = bibliotekar.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Baza znanja = biblioteka.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Stanje = ono što kontrolna tabla pokazuje upravo sada.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Alati = ruke koje zaista mogu nešto da urade.\u003C\u002Fstrong>\u003C\u002Fdiv>\u003C\u002Faside>\n\u003Ch2 id=\"section-75\">Zaključak\u003C\u002Fh2>\n\u003Cp>RAG je mnogo manje misteriozan kada se delovi razdvoje.\u003C\u002Fp>\n\u003Cp>LLM razume i generiše jezik. Aplikacija održava trenutno stanje. Baza znanja čuva informacije. RAG pronalazi korisni deo tih informacija i stavlja ga u kontekst LLM-a. Alati ili aplikacija obavljaju stvarne radnje.\u003C\u002Fp>\n\u003Cp>To je osnovna arhitektura koja stoji iza mnogih modernih AI asistenata i agenata.\u003C\u002Fp>\n\u003Ch2 id=\"section-79\">Često postavljana pitanja\u003C\u002Fh2>\n\u003Csection class=\"editorjs-faq my-6 rounded-xl border border-gray-200 p-5 dark:border-gray-700\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">RAG jednostavnim rečima\u003C\u002Fh3>\u003Cdiv id=\"faq1\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Šta je RAG jednostavno rečeno?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">RAG je korak u kojem AI pretražuje izvor znanja radi relevantnih informacija pre nego što jezički model napiše svoj odgovor.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq2\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je RAG-u potreban internet?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne. Baza znanja može biti potpuno lokalna na vašem računaru ili serveru.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq3\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je RAG isto što i baza podataka?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne. Baza podataka ili datoteke sadrže informacije. RAG je proces pronalaženja koji pronalazi korisni deo i daje ga LLM-u.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq4\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je RAG isto što i memorija?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne. Memorija obično čuva prethodne interakcije ili događaje. RAG pronalazi relevantno znanje kada je potrebno.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq5\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li je trenutno stanje aplikacije deo RAG-a?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne nužno. Trenutno stanje se obično dobija direktno iz aplikacije ili skladišta stanja. RAG je bolje razumeti kao pronalaženje iz izvora znanja.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003Cdiv id=\"faq6\" class=\"border-t border-gray-200 py-4 first:border-t-0 dark:border-gray-700\">\u003Ch4 class=\"font-semibold text-gray-900 dark:text-gray-100\">Da li RAG čini AI odgovore tačnim?\u003C\u002Fh4>\u003Cdiv class=\"mt-2 text-gray-600 dark:text-gray-300\">Ne. Može pružiti bolje dokaze, ali pronalaženje i dalje može biti pogrešno ili zastarelo, a LLM i dalje može pogrešno da zaključuje.\u003C\u002Fdiv>\u003C\u002Fdiv>\u003C\u002Fsection>\n\u003Ch2 id=\"section-81\">Pojmovnik\u003C\u002Fh2>\n\u003Csection class=\"editorjs-glossary my-6 rounded-xl border border-gray-200 dark:border-gray-700 p-5\">\u003Ch3 class=\"mb-3 text-lg font-semibold\">Osnovni pojmovi\u003C\u002Fh3>\u003Cdl>\u003Cdiv id=\"llm\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">LLM\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Jezički model koji razume i generiše tekst i može da zaključuje na osnovu informacija stavljenih u njegov kontekst.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"rag\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">RAG\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Retrieval-Augmented Generation: pronalaženje relevantnih spoljnih informacija i njihovo dodavanje u kontekst modela pre generisanja odgovora.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"knowledge-base\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Baza znanja\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Datoteke, dokumenti, zapisi ili druge informacije koje pretraga može da pretražuje.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"state\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Stanje\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Trenutne činjenice aplikacije, sistema ili sveta u određenom trenutku.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"context\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Kontekst\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Informacije koje se trenutno dostavljaju jezičkom modelu za jedan zahtev ili korak zaključivanja.\u003C\u002Fdd>\u003C\u002Fdiv>\u003Cdiv id=\"embedding\" class=\"border-t border-gray-200 dark:border-gray-700 py-3 first:border-t-0\">\u003Cdt class=\"font-semibold text-gray-900 dark:text-gray-100\">Embedding\u003C\u002Fdt>\u003Cdd class=\"mt-1 text-gray-600 dark:text-gray-300\">Numerička reprezentacija značenja koja može pomoći semantičkoj pretrazi da pronađe konceptualno slične informacije.\u003C\u002Fdd>\u003C\u002Fdiv>\u003C\u002Fdl>\u003C\u002Fsection>\n\u003Ch2 id=\"section-83\">Primarni izvori\u003C\u002Fh2>\n\u003Ca href=\"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fapi-reference\u002Fvector-stores-files\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">OpenAI — Datoteke vektorske memorije\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Zvanična dokumentacija koja pokazuje kako se datoteke mogu priložiti vektorskim skladištima, podeliti na delove i učiniti dostupnim za pretragu datoteka.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fquickstart\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">OpenAI — Brzi početak za programere\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Zvanična OpenAI dokumentacija koja opisuje alate kao što je pretraga datoteka za davanje modelima pristupa spoljnim informacijama.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdeveloper.nvidia.com\u002Face-for-games\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NVIDIA Developer — ACE za igre\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Zvanična NVIDIA dokumentacija koja opisuje odvojene Agent, Chat i RAG API-je za povezivanje likova u igrama sa stanjem igre, kontekstualnim znanjem i radnjama vođenim modelom.\u003C\u002Fp>\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdeveloper.nvidia.com\u002Fblog\u002Fhow-krafton-built-pubg-ally-a-co-playable-character-powered-by-nvidia-ace\u002F\" target=\"_blank\" rel=\"noopener noreferrer\" class=\"editorjs-link-tool block border border-gray-200 dark:border-gray-700 rounded-lg p-4 transition text-gray-900 dark:text-gray-100 hover:border-primary-500 hover:bg-primary-50 dark:hover:bg-gray-900 hover:text-gray-900 dark:hover:text-gray-100\">\u003Cstrong class=\"block font-semibold\">NVIDIA Developer — Kako je KRAFTON napravio PUBG Ally\u003C\u002Fstrong>\u003Cp class=\"text-sm text-gray-600 dark:text-gray-400\">Zvanično tehničko objašnjenje koje razdvaja stanje uživo meča od pretrage znanja i zaključivanja jezičkog modela.\u003C\u002Fp>\u003C\u002Fa>",{"time":212,"blocks":213,"version":873},1790377539809,[214,220,228,233,241,246,251,256,261,268,273,278,283,288,295,300,320,325,330,335,340,361,366,371,376,381,386,391,422,429,434,439,444,449,454,459,464,485,490,496,501,506,511,516,521,526,531,536,541,546,551,556,561,590,595,600,605,610,615,624,629,634,655,660,665,670,675,680,685,690,695,700,705,741,747,752,757,762,767,772,802,807,831,836,846,855,864],{"id":215,"data":216,"type":218,"tunes":219},"intro",{"text":217},"RAG zvuči komplikovano jer je ime komplikovano. Ideja nije. RAG jednostavno znači: pre nego što AI odgovori, prvo pronađe relevantne informacije iz izvora znanja i daje te informacije jezičkom modelu.","paragraph",{},{"id":221,"data":222,"type":226,"tunes":227},"one-sentence",{"body":223,"title":224,"variant":225},"\u003Cstrong>RAG je korak u kojem AI pretražuje bazu znanja za korisne informacije pre nego što LLM napiše odgovor.\u003C\u002Fstrong>","RAG u jednoj rečenici","info","callout",{},{"id":229,"data":230,"type":218,"tunes":232},"analogy",{"text":231},"Zamislite LLM kao pametnu osobu koja sedi za stolom. RAG je bibliotekar koji donosi pravu stranicu iz prave knjige. LLM zatim čita tu stranicu i odgovara vam.",{},{"id":234,"data":235,"type":239,"tunes":240},"toc",{"title":236,"maxLevel":237,"minLevel":238},"Sadržaj",3,2,"tableOfContents",{},{"id":242,"data":243,"type":42,"tunes":245},"h-llm",{"text":244,"level":238},"Prvo: šta radi LLM?",{},{"id":247,"data":248,"type":218,"tunes":250},"p-llm-1",{"text":249},"LLM je deo koji razume jezik i proizvodi jezik. Može da pročita vaše pitanje, razume instrukcije, uporedi informacije, objasni nešto i napiše odgovor.",{},{"id":252,"data":253,"type":218,"tunes":255},"p-llm-2",{"text":254},"Ali LLM ne zna automatski šta se trenutno nalazi u bazi podataka vaše kompanije, vašoj igračkoj sesiji, vašim privatnim dokumentima ili datoteci koju ste kreirali pre pet minuta.",{},{"id":257,"data":258,"type":218,"tunes":260},"p-llm-3",{"text":259},"Zna samo ono što je već unutar modela plus sve informacije koje mu aplikacija pruži u trenutnom zahtevu.",{},{"id":262,"data":263,"type":226,"tunes":267},"llm-rule",{"body":264,"title":265,"variant":266},"LLM \u003Cstrong>misli i piše\u003C\u002Fstrong>. Ne poseduje automatski sve vaše trenutne podatke.","Jednostavno pravilo","note",{},{"id":269,"data":270,"type":42,"tunes":272},"h-kb",{"text":271,"level":238},"Zatim: šta je baza znanja?",{},{"id":274,"data":275,"type":218,"tunes":277},"p-kb-1",{"text":276},"Baza znanja je jednostavno informacija koju aplikacija može da pretražuje.",{},{"id":279,"data":280,"type":218,"tunes":282},"p-kb-2",{"text":281},"Može da sadrži PDF-ove, priručnike, dokumentaciju proizvoda, članke podrške, ugovore, pravila igre, podatke o oružju, interne dokumente kompanije, zapise baze podataka ili drugi tekst.",{},{"id":284,"data":285,"type":218,"tunes":287},"p-kb-3",{"text":286},"Baza znanja može biti lokalna na vašem računaru. Može biti na serveru. Može biti u vektorskoj bazi podataka. Takođe može biti izgrađena od običnih datoteka. RAG ne znači Internet.",{},{"id":289,"data":290,"type":226,"tunes":294},"no-internet",{"body":291,"title":292,"variant":293},"\u003Cstrong>RAG ne zahteva Internet.\u003C\u002Fstrong> Informacije mogu biti potpuno lokalne.","Važno","success",{},{"id":296,"data":297,"type":42,"tunes":299},"h-rag",{"text":298,"level":238},"Dakle, šta RAG zapravo radi?",{},{"id":301,"data":302,"type":318,"tunes":319},"rag-flow",{"steps":303,"title":316,"orientation":317},[304,307,310,313],{"label":305,"description":306},"1. Postavljate pitanje","Na primer: Koju municiju koristi ovo oružje?",{"label":308,"description":309},"2. RAG pretražuje bazu znanja","Sistem traži male delove informacija koji su najrelevantniji za vaše pitanje.",{"label":311,"description":312},"3. RAG daje te delove LLM-u","LLM prima pitanje plus pronađene informacije.",{"label":314,"description":315},"4. LLM piše odgovor","Koristi pronađene informacije kao kontekst za odgovor.","Ceo RAG proces","auto","processFlow",{},{"id":321,"data":322,"type":218,"tunes":324},"rag-that-is-it",{"text":323},"To je RAG.",{},{"id":326,"data":327,"type":218,"tunes":329},"rag-name",{"text":328},"Puno ime je Retrieval-Augmented Generation. Retrieval znači pronalaženje relevantnih informacija. Augmented znači dodavanje tih informacija u kontekst modela. Generation znači da LLM piše konačni odgovor.",{},{"id":331,"data":332,"type":42,"tunes":334},"h-example",{"text":333,"level":238},"Vrlo jednostavan primer",{},{"id":336,"data":337,"type":218,"tunes":339},"p-ex-1",{"text":338},"Zamislite da imate lokalnu bazu znanja o igri.",{},{"id":341,"data":342,"type":359,"tunes":360},"kb-table",{"content":343,"stretched":43,"withHeadings":14},[344,347,350,353,356],[345,346],"Baza znanja sadrži","Primer",[348,349],"Oružja","AKM koristi municiju 7.62 mm",[351,352],"Predmeti za lečenje","Med Kit obnavlja zdravlje",[354,355],"Dodaci","Ovaj dodatak radi sa ovim oružjima",[357,358],"Pravila mape","Ova zona se ponaša na ovaj način","table",{},{"id":362,"data":363,"type":218,"tunes":365},"p-ex-2",{"text":364},"Pitate: „Koju municiju koristi AKM?“",{},{"id":367,"data":368,"type":218,"tunes":370},"p-ex-3",{"text":369},"RAG pretražuje bazu znanja i pronalazi unos o AKM. Taj mali deo informacije daje LLM-u. LLM zatim odgovara: „AKM koristi municiju 7.62 mm.“",{},{"id":372,"data":373,"type":218,"tunes":375},"p-ex-4",{"text":374},"LLM nije bio potreban cela baza podataka. RAG je doneo samo koristan deo.",{},{"id":377,"data":378,"type":42,"tunes":380},"h-state",{"text":379,"level":238},"Sada važan deo: RAG nije trenutno stanje",{},{"id":382,"data":383,"type":218,"tunes":385},"p-state-1",{"text":384},"Ovde mnoga objašnjenja postaju zbunjujuća.",{},{"id":387,"data":388,"type":218,"tunes":390},"p-state-2",{"text":389},"RAG obično daje AI znanje. Sistem stanja daje AI činjenice o tome šta je istinito upravo sada.",{},{"id":392,"data":393,"type":420,"tunes":421},"knowledge-state",{"rows":394,"title":412,"layout":359,"columns":413},[395,400,404,408],{"id":396,"label":397,"values":398},"weapon","Oružje",[399,399],"",{"id":401,"label":402,"values":403},"ammo","Municija",[399,399],{"id":405,"label":406,"values":407},"health","Zdravlje",[399,399],{"id":409,"label":410,"values":411},"enemy","Neprijatelj",[399,399],"Znanje naspram trenutnog stanja",[414,417],{"id":415,"label":416},"knowledge","RAG \u002F znanje",{"id":418,"label":419},"state","Trenutno stanje","comparison",{},{"id":423,"data":424,"type":226,"tunes":428},"dont-mix",{"body":425,"title":426,"variant":427},"RAG odgovara: \u003Cstrong>Šta je generalno istinito?\u003C\u002Fstrong>\u003Cbr>Stanje odgovara: \u003Cstrong>Šta je istinito upravo sada?\u003C\u002Fstrong>","Ne mešajte ovo dvoje","warning",{},{"id":430,"data":431,"type":42,"tunes":433},"h-state-db",{"text":432,"level":238},"Šta je baza podataka stanja?",{},{"id":435,"data":436,"type":218,"tunes":438},"p-statedb-1",{"text":437},"Baza podataka stanja ili skladište stanja je jednostavno mesto gde aplikacija čuva trenutne činjenice.",{},{"id":440,"data":441,"type":218,"tunes":443},"p-statedb-2",{"text":442},"U igri, engine već zna stvari kao što su vaše zdravlje, pozicija, inventar, municija, trenutna misija, obližnji objekti i status neprijatelja. AI sistem može izložiti odabrane delove tog stanja modelu.",{},{"id":445,"data":446,"type":218,"tunes":448},"p-statedb-3",{"text":447},"U poslovnoj aplikaciji, ista ideja može biti baza podataka narudžbina, zapis o kupcu, status projekta ili trenutna vrednost senzora.",{},{"id":450,"data":451,"type":218,"tunes":453},"p-statedb-4",{"text":452},"Stanje kreira sama aplikacija dok se stvari dešavaju. Ako izgubite zdravlje, igra ažurira vrednost zdravlja. Ako pokupite municiju, inventar se menja. Ako je narudžbina plaćena, poslovni sistem menja status narudžbine.",{},{"id":455,"data":456,"type":226,"tunes":458},"state-rule",{"body":457,"title":265,"variant":225},"Aplikacija kreira i ažurira \u003Cstrong>stanje\u003C\u002Fstrong>. RAG pretražuje \u003Cstrong>znanje\u003C\u002Fstrong>. LLM koristi oboje da odluči šta da kaže ili uradi.",{},{"id":460,"data":461,"type":42,"tunes":463},"h-together",{"text":462,"level":238},"Kako tri dela rade zajedno",{},{"id":465,"data":466,"type":318,"tunes":484},"together-flow",{"steps":467,"title":483,"orientation":317},[468,471,474,477,480],{"label":469,"description":470},"1. Trenutno stanje","Aplikacija govori AI-ju šta je sada istina: zdravlje 41%, AKM opremljen, 23 metka.",{"label":472,"description":473},"2. RAG","Sistem pronalazi korisno znanje: kako oružje funkcioniše, koji predmet za lečenje je dostupan ili relevantno pravilo.",{"label":475,"description":476},"3. LLM","Model prima pitanje, trenutno stanje i pronađeno znanje.",{"label":478,"description":479},"4. Rezonovanje","LLM kombinuje te ulaze i odlučuje koji odgovor ili akcija na visokom nivou ima smisla.",{"label":481,"description":482},"5. Aplikacija","Ako je potrebna akcija, aplikacija ili game engine je izvršava i ponovo ažurira stanje.","LLM + stanje + RAG",{},{"id":486,"data":487,"type":218,"tunes":489},"p-arch-intro",{"text":488},"Dakle, osnovna arhitektura je:",{},{"id":491,"data":492,"type":226,"tunes":495},"simple-architecture",{"body":493,"title":494,"variant":266},"\u003Cstrong>Stanje = šta je sada istina\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = korisno znanje\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>LLM = razume, rezonuje i piše\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Aplikacija = izvršava pravu akciju\u003C\u002Fstrong>","Najjednostavnija arhitektura",{},{"id":497,"data":498,"type":42,"tunes":500},"h-vector",{"text":499,"level":238},"Da li RAG uvek koristi vektorsku bazu podataka?",{},{"id":502,"data":503,"type":218,"tunes":505},"p-vector-1",{"text":504},"Ne.",{},{"id":507,"data":508,"type":218,"tunes":510},"p-vector-2",{"text":509},"Vektorska baza podataka je uobičajen način za izgradnju semantičke pretrage, ali to nije definicija RAG-a.",{},{"id":512,"data":513,"type":218,"tunes":515},"p-vector-3",{"text":514},"Važan deo je pronalaženje: sistem pronalazi relevantne spoljne informacije i dodaje ih u kontekst LLM-a pre nego što se odgovor generiše.",{},{"id":517,"data":518,"type":218,"tunes":520},"p-vector-4",{"text":519},"OpenAI-jev File Search, na primer, može da radi sa fajlovima sačuvanim u vektorskim skladištima. Fajlovi se dele na manje delove kako bi sistem mogao da pronađe delove koji su relevantni za pitanje. To je jedna implementacija iste osnovne ideje.",{},{"id":522,"data":523,"type":42,"tunes":525},"h-embedding",{"text":524,"level":238},"Šta je embedding, jednostavno rečeno?",{},{"id":527,"data":528,"type":218,"tunes":530},"p-emb-1",{"text":529},"Ne morate da razumete embedding-e da biste razumeli RAG.",{},{"id":532,"data":533,"type":218,"tunes":535},"p-emb-2",{"text":534},"Ali jednostavna verzija je ovo: embedding je numerička reprezentacija značenja. Pomaže sistemu za pretragu da pronađe tekst koji je konceptualno sličan čak i kada reči nisu potpuno iste.",{},{"id":537,"data":538,"type":218,"tunes":540},"p-emb-3",{"text":539},"Na primer, obična pretraga po ključnim rečima može da traži tačne reči „popravka automobila“. Semantička pretraga takođe može da razume da se „popravi moje vozilo“ odnosi na sličnu temu.",{},{"id":542,"data":543,"type":218,"tunes":545},"p-emb-4",{"text":544},"To čini embedding-e korisnim za RAG, ali RAG takođe može da koristi pretragu po ključnim rečima, upite baze podataka ili hibrid nekoliko metoda.",{},{"id":547,"data":548,"type":42,"tunes":550},"h-memory",{"text":549,"level":238},"RAG takođe nije memorija",{},{"id":552,"data":553,"type":218,"tunes":555},"p-memory-1",{"text":554},"Memorija je još jedan koncept koji se često meša sa RAG-om.",{},{"id":557,"data":558,"type":218,"tunes":560},"p-memory-2",{"text":559},"Memorija je obično informacija koju sistem čuva o prethodnim interakcijama ili prethodnim događajima. RAG je mehanizam koji se koristi za pronalaženje relevantnog znanja kada je potrebno.",{},{"id":562,"data":563,"type":359,"tunes":589},"parts-table",{"content":564,"stretched":43,"withHeadings":14},[565,568,571,574,577,580,583,586],[566,567],"Deo","Jednostavno značenje",[569,570],"LLM","Deo koji razume i generiše jezik",[572,573],"RAG","Deo koji traži relevantno znanje pre odgovora",[575,576],"Baza znanja","Informacije koje RAG može da pretražuje",[578,579],"Stanje","Šta je trenutno istina u aplikaciji ili svetu",[581,582],"Memorija","Informacije sačuvane iz prethodnih interakcija ili događaja",[584,585],"Alat \u002F akcija","Nešto što AI sme da pozove ili zatraži od aplikacije da uradi",[587,588],"Kontekst","Informacije trenutno stavljene pred LLM za ovaj zahtev",{},{"id":591,"data":592,"type":42,"tunes":594},"h-pubg",{"text":593,"level":238},"Pravi primer iz igre: PUBG Ally",{},{"id":596,"data":597,"type":218,"tunes":599},"p-pubg-1",{"text":598},"PUBG Ally je koristan primer jer čini razliku vidljivom.",{},{"id":601,"data":602,"type":218,"tunes":604},"p-pubg-2",{"text":603},"KRAFTON opisuje stanje meča uživo kao poseban izvor istine. Igra izlaže trenutne činjenice kroz alate za posmatranje: trenutno oružje, municiju, zdravlje, status sigurne zone, obližnje predmete i borbenu situaciju.",{},{"id":606,"data":607,"type":218,"tunes":609},"p-pubg-3",{"text":608},"Pretraga znanja je drugačiji posao. Sistem može da koristi pripremljeno znanje o oružju, dodacima, predmetima i pravilima. NVIDIA ACE Game Agent SDK takođe izlaže poseban RAG API za pronalaženje znanja iz baza koje su napravili programeri.",{},{"id":611,"data":612,"type":218,"tunes":614},"p-pubg-4",{"text":613},"To nam daje jasnu podelu: game engine kaže šta se sada dešava, pretraga obezbeđuje relevantno znanje, a jezički model odlučuje šta informacije znače.",{},{"id":616,"data":617,"type":622,"tunes":623},"ref-pubg",{"url":618,"title":619,"excerpt":620,"ctaLabel":621},"https:\u002F\u002Ffigure.rocks\u002Fblog\u002Fpubg-ally-shows-why-ai-teammates-need-two-brains-fast-reflexes-and-slow-reasoning","PUBG Ally pokazuje zašto AI saigračima trebaju dva mozga: brzi refleksi i sporo rezonovanje","Praktičan primer iz igre koji pokazuje kako stanje uživo, jezičko rezonovanje i deterministička kontrola na strani igre mogu da rade zajedno.","Pročitajte članak o arhitekturi PUBG Ally","referralArticle",{},{"id":625,"data":626,"type":42,"tunes":628},"h-complete",{"text":627,"level":238},"Jedan kompletan primer",{},{"id":630,"data":631,"type":218,"tunes":633},"p-complete-1",{"text":632},"Zamislite da kažete AI saigraču: „Nisko mi je zdravlje. Treba li da napadnemo?“",{},{"id":635,"data":636,"type":318,"tunes":654},"complete-flow",{"steps":637,"title":653,"orientation":317},[638,640,642,644,647,650],{"label":578,"description":639},"Igra prijavljuje: zdravlje 24%, jedan neprijatelj u blizini, dva predmeta za lečenje dostupna.",{"label":572,"description":641},"Sistem znanja pronalazi relevantna pravila za predmet za lečenje i možda informacije o trenutnom oružju ili taktičkom mehanizmu.",{"label":569,"description":643},"Model kombinuje vaš zahtev, trenutno stanje i pronađeno znanje.",{"label":645,"description":646},"Odluka","Zaključuje da je lečenje prvo sigurnije od trenutnog napada.",{"label":648,"description":649},"Alat \u002F game engine","Agent zahteva legalnu akciju u igri kao što je premeštanje u zaklon ili korišćenje predmeta za lečenje.",{"label":651,"description":652},"Novo stanje","Igra izvršava akciju i prijavljuje ažuriranu situaciju nazad agentu.","Šta se dešava dalje",{},{"id":656,"data":657,"type":218,"tunes":659},"p-complete-2",{"text":658},"RAG nije kontrolisao lika. Baza podataka o stanju nije rezonovala. LLM nije direktno menjao igru. Svaki deo je imao jedan zadatak.",{},{"id":661,"data":662,"type":42,"tunes":664},"h-why",{"text":663,"level":238},"Zašto uopšte koristiti RAG?",{},{"id":666,"data":667,"type":218,"tunes":669},"p-why-1",{"text":668},"Zato što bi stavljanje svakog dokumenta, pravila i zapisa iz baze u svaki prompt bilo sporo, skupo i često zbunjujuće.",{},{"id":671,"data":672,"type":218,"tunes":674},"p-why-2",{"text":673},"RAG omogućava sistemu da izabere samo informacije koje su korisne za trenutno pitanje.",{},{"id":676,"data":677,"type":218,"tunes":679},"p-why-3",{"text":678},"Takođe vam omogućava da ažurirate bazu znanja bez ponovnog treniranja celog jezičkog modela. Promenite dokument ili bazu podataka, ponovo izgradite ili osvežite indeks kada je potrebno, i sledeća pretraga može da koristi novije informacije.",{},{"id":681,"data":682,"type":42,"tunes":684},"h-not-guarantee",{"text":683,"level":238},"Šta RAG ne garantuje",{},{"id":686,"data":687,"type":218,"tunes":689},"p-not-1",{"text":688},"RAG može da poboljša utemeljenje, ali ne čini odgovor automatski tačnim.",{},{"id":691,"data":692,"type":218,"tunes":694},"p-not-2",{"text":693},"Korak pretrage može da pronađe pogrešan dokument. Tačan dokument može biti zastareo. LLM može da pogrešno razume dobre dokaze. Ili se trenutno stanje može promeniti.",{},{"id":696,"data":697,"type":218,"tunes":699},"p-not-3",{"text":698},"Pouzdan sistem stoga mora odvojeno da validira pretragu, svežinu stanja i konačno rezonovanje modela.",{},{"id":701,"data":702,"type":42,"tunes":704},"h-mental",{"text":703,"level":238},"Najlakši mentalni model za pamćenje",{},{"id":706,"data":707,"type":420,"tunes":740},"mental-table",{"rows":708,"title":733,"layout":359,"columns":734},[709,713,717,721,725,729],{"id":710,"label":711,"values":712},"brain","Osoba koja razmišlja",[399,399],{"id":714,"label":715,"values":716},"library","Pronalaženje priručnika",[399,399],{"id":718,"label":719,"values":720},"books","Knjige na polici",[399,399],{"id":722,"label":723,"values":724},"dashboard","Trenutna kontrolna tabla ili instrument tabla",[399,399],{"id":726,"label":727,"values":728},"notes","Beleške sa ranijih sastanaka",[399,399],{"id":730,"label":731,"values":732},"hands","Radi nešto u stvarnom svetu",[399,399],"Zamislite AI sistem kao osobu za radnim stolom",[735,737],{"id":229,"label":736},"Analogija",{"id":738,"label":739},"system","AI sistem",{},{"id":742,"data":743,"type":226,"tunes":746},"remember",{"body":744,"title":745,"variant":293},"\u003Cstrong>LLM = mozak.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = bibliotekar.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Baza znanja = biblioteka.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Stanje = ono što kontrolna tabla pokazuje upravo sada.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Alati = ruke koje zaista mogu nešto da urade.\u003C\u002Fstrong>","Ako zapamtite samo ovo",{},{"id":748,"data":749,"type":42,"tunes":751},"h-conclusion",{"text":750,"level":238},"Zaključak",{},{"id":753,"data":754,"type":218,"tunes":756},"p-conc-1",{"text":755},"RAG je mnogo manje misteriozan kada se delovi razdvoje.",{},{"id":758,"data":759,"type":218,"tunes":761},"p-conc-2",{"text":760},"LLM razume i generiše jezik. Aplikacija održava trenutno stanje. Baza znanja čuva informacije. RAG pronalazi korisni deo tih informacija i stavlja ga u kontekst LLM-a. Alati ili aplikacija obavljaju stvarne radnje.",{},{"id":763,"data":764,"type":218,"tunes":766},"p-conc-3",{"text":765},"To je osnovna arhitektura koja stoji iza mnogih modernih AI asistenata i agenata.",{},{"id":768,"data":769,"type":42,"tunes":771},"h-faq",{"text":770,"level":238},"Često postavljana pitanja",{},{"id":773,"data":774,"type":773,"tunes":801},"faq",{"items":775,"title":800},[776,780,784,788,792,796],{"id":777,"answer":778,"question":779},"faq1","RAG je korak u kojem AI pretražuje izvor znanja radi relevantnih informacija pre nego što jezički model napiše svoj odgovor.","Šta je RAG jednostavno rečeno?",{"id":781,"answer":782,"question":783},"faq2","Ne. Baza znanja može biti potpuno lokalna na vašem računaru ili serveru.","Da li je RAG-u potreban internet?",{"id":785,"answer":786,"question":787},"faq3","Ne. Baza podataka ili datoteke sadrže informacije. RAG je proces pronalaženja koji pronalazi korisni deo i daje ga LLM-u.","Da li je RAG isto što i baza podataka?",{"id":789,"answer":790,"question":791},"faq4","Ne. Memorija obično čuva prethodne interakcije ili događaje. RAG pronalazi relevantno znanje kada je potrebno.","Da li je RAG isto što i memorija?",{"id":793,"answer":794,"question":795},"faq5","Ne nužno. Trenutno stanje se obično dobija direktno iz aplikacije ili skladišta stanja. RAG je bolje razumeti kao pronalaženje iz izvora znanja.","Da li je trenutno stanje aplikacije deo RAG-a?",{"id":797,"answer":798,"question":799},"faq6","Ne. Može pružiti bolje dokaze, ali pronalaženje i dalje može biti pogrešno ili zastarelo, a LLM i dalje može pogrešno da zaključuje.","Da li RAG čini AI odgovore tačnim?","RAG jednostavnim rečima",{},{"id":803,"data":804,"type":42,"tunes":806},"h-glossary",{"text":805,"level":238},"Pojmovnik",{},{"id":808,"data":809,"type":808,"tunes":830},"glossary",{"title":810,"entries":811},"Osnovni pojmovi",[812,815,818,821,823,826],{"term":569,"anchor":813,"definition":814},"llm","Jezički model koji razume i generiše tekst i može da zaključuje na osnovu informacija stavljenih u njegov kontekst.",{"term":572,"anchor":816,"definition":817},"rag","Retrieval-Augmented Generation: pronalaženje relevantnih spoljnih informacija i njihovo dodavanje u kontekst modela pre generisanja odgovora.",{"term":575,"anchor":819,"definition":820},"knowledge-base","Datoteke, dokumenti, zapisi ili druge informacije koje pretraga može da pretražuje.",{"term":578,"anchor":418,"definition":822},"Trenutne činjenice aplikacije, sistema ili sveta u određenom trenutku.",{"term":587,"anchor":824,"definition":825},"context","Informacije koje se trenutno dostavljaju jezičkom modelu za jedan zahtev ili korak zaključivanja.",{"term":827,"anchor":828,"definition":829},"Embedding","embedding","Numerička reprezentacija značenja koja može pomoći semantičkoj pretrazi da pronađe konceptualno slične informacije.",{},{"id":832,"data":833,"type":42,"tunes":835},"h-sources",{"text":834,"level":238},"Primarni izvori",{},{"id":837,"data":838,"type":844,"tunes":845},"src-openai-vector",{"link":839,"meta":840},"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fapi-reference\u002Fvector-stores-files",{"image":841,"title":842,"description":843},{"url":399},"OpenAI — Datoteke vektorske memorije","Zvanična dokumentacija koja pokazuje kako se datoteke mogu priložiti vektorskim skladištima, podeliti na delove i učiniti dostupnim za pretragu datoteka.","linkTool",{},{"id":847,"data":848,"type":844,"tunes":854},"src-openai-quickstart",{"link":849,"meta":850},"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fquickstart",{"image":851,"title":852,"description":853},{"url":399},"OpenAI — Brzi početak za programere","Zvanična OpenAI dokumentacija koja opisuje alate kao što je pretraga datoteka za davanje modelima pristupa spoljnim informacijama.",{},{"id":856,"data":857,"type":844,"tunes":863},"src-nvidia-ace",{"link":858,"meta":859},"https:\u002F\u002Fdeveloper.nvidia.com\u002Face-for-games",{"image":860,"title":861,"description":862},{"url":399},"NVIDIA Developer — ACE za igre","Zvanična NVIDIA dokumentacija koja opisuje odvojene Agent, Chat i RAG API-je za povezivanje likova u igrama sa stanjem igre, kontekstualnim znanjem i radnjama vođenim modelom.",{},{"id":865,"data":866,"type":844,"tunes":872},"src-nvidia-pubg",{"link":867,"meta":868},"https:\u002F\u002Fdeveloper.nvidia.com\u002Fblog\u002Fhow-krafton-built-pubg-ally-a-co-playable-character-powered-by-nvidia-ace\u002F",{"image":869,"title":870,"description":871},{"url":399},"NVIDIA Developer — Kako je KRAFTON napravio PUBG Ally","Zvanično tehničko objašnjenje koje razdvaja stanje uživo meča od pretrage znanja i zaključivanja jezičkog modela.",{},"2.31","RAG zvuči komplikovano, ali ideja je jednostavna: pre nego što AI odgovori, prvo potraži korisne informacije iz izvora znanja i daje te informacije jezičkom modelu. Ovaj vodič objašnjava RAG, LLM-ove, stanje, memoriju i alate koristeći jedan jednostavan mentalni model.","\u002Fuploads\u002F2026\u002F09\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works-1790377492124-khjagt.webp","what-is-rag-the-simplest-explanation-of-how-it-works-1790377492124-khjagt","PUBLISHED","2026-09-25T19:03:00.000Z","2026-09-25T23:03:13.651Z","2026-09-25T23:41:17.272Z",{"en":882,"de":883,"sr":884,"es":885,"fr":886,"it":887,"ru":888,"zh":889},"\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fde\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fsr\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fes\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Ffr\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fit\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fru\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works","\u002Fzh\u002Fblog\u002Fwhat-is-rag-the-simplest-explanation-of-how-it-works",[891,895,899,903,907,911],{"id":892,"name":893,"slug":894},46,"Pregled","overview",{"id":896,"name":897,"slug":898},57,"Granice podataka","data-boundaries",{"id":900,"name":901,"slug":902},51,"Anti-obrasci","anti-patterns",{"id":904,"name":905,"slug":906},58,"Evaluacija i gate-ovi kvaliteta","evaluation",{"id":908,"name":909,"slug":910},56,"Portfolio slučajeva upotrebe","use-case-portfolio",{"id":912,"name":913,"slug":914},60,"Kontrole troška i latencije","cost-and-latency",{"id":916,"login":917,"email":918,"displayName":919},"20","rooth8233","aleksandar@stajic.de","Aleksandar Stajić",[921,1452],{"lang":922,"title":923,"content":924,"contentJson":925,"excerpt":1451},"en","What Is RAG? The Simplest Explanation of How It Works","{\"time\":1790377494031,\"blocks\":[{\"id\":\"intro\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG sounds complicated because the name is complicated. The idea is not. RAG simply means: before the AI answers, it first looks up relevant information from a knowledge source and gives that information to the language model.\"},\"tunes\":{}},{\"id\":\"one-sentence\",\"type\":\"callout\",\"data\":{\"variant\":\"info\",\"title\":\"RAG in one sentence\",\"body\":\"\u003Cstrong>RAG is the step where an AI searches a knowledge base for useful information before the LLM writes the answer.\u003C\u002Fstrong>\"},\"tunes\":{}},{\"id\":\"analogy\",\"type\":\"paragraph\",\"data\":{\"text\":\"Think of an LLM as a smart person sitting at a desk. RAG is the librarian who brings the right page from the right book. The LLM then reads that page and answers you.\"},\"tunes\":{}},{\"id\":\"toc\",\"type\":\"tableOfContents\",\"data\":{\"title\":\"Contents\",\"minLevel\":2,\"maxLevel\":3},\"tunes\":{}},{\"id\":\"h-llm\",\"type\":\"header\",\"data\":{\"text\":\"First: what does the LLM do?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-llm-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"The LLM is the part that understands language and produces language. It can read your question, understand instructions, compare information, explain something and write an answer.\"},\"tunes\":{}},{\"id\":\"p-llm-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"But the LLM does not automatically know what is currently inside your company database, your game session, your private documents or a file you created five minutes ago.\"},\"tunes\":{}},{\"id\":\"p-llm-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"It only knows what is already inside the model plus whatever information the application gives it in the current request.\"},\"tunes\":{}},{\"id\":\"llm-rule\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"Simple rule\",\"body\":\"The LLM \u003Cstrong>thinks and writes\u003C\u002Fstrong>. It does not automatically own all of your current data.\"},\"tunes\":{}},{\"id\":\"h-kb\",\"type\":\"header\",\"data\":{\"text\":\"Then: what is the knowledge base?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-kb-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A knowledge base is simply information the application can search.\"},\"tunes\":{}},{\"id\":\"p-kb-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"It could contain PDFs, manuals, product documentation, support articles, contracts, game rules, weapon data, internal company documents, database records or other text.\"},\"tunes\":{}},{\"id\":\"p-kb-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"The knowledge base can be local on your own machine. It can be on a server. It can be in a vector database. It can also be built from normal files. RAG does not mean Internet.\"},\"tunes\":{}},{\"id\":\"no-internet\",\"type\":\"callout\",\"data\":{\"variant\":\"success\",\"title\":\"Important\",\"body\":\"\u003Cstrong>RAG does not require the Internet.\u003C\u002Fstrong> The information can be completely local.\"},\"tunes\":{}},{\"id\":\"h-rag\",\"type\":\"header\",\"data\":{\"text\":\"So what does RAG actually do?\",\"level\":2},\"tunes\":{}},{\"id\":\"rag-flow\",\"type\":\"processFlow\",\"data\":{\"title\":\"The whole RAG process\",\"orientation\":\"auto\",\"steps\":[{\"label\":\"1. You ask a question\",\"description\":\"For example: Which ammunition does this weapon use?\"},{\"label\":\"2. RAG searches the knowledge base\",\"description\":\"The system looks for the small pieces of information most relevant to your question.\"},{\"label\":\"3. RAG gives those pieces to the LLM\",\"description\":\"The LLM receives the question plus the retrieved information.\"},{\"label\":\"4. The LLM writes the answer\",\"description\":\"It uses the retrieved information as context for the response.\"}]},\"tunes\":{}},{\"id\":\"rag-that-is-it\",\"type\":\"paragraph\",\"data\":{\"text\":\"That is RAG.\"},\"tunes\":{}},{\"id\":\"rag-name\",\"type\":\"paragraph\",\"data\":{\"text\":\"The full name is Retrieval-Augmented Generation. Retrieval means finding the relevant information. Augmented means adding that information to the model's context. Generation means the LLM writes the final answer.\"},\"tunes\":{}},{\"id\":\"h-example\",\"type\":\"header\",\"data\":{\"text\":\"A very simple example\",\"level\":2},\"tunes\":{}},{\"id\":\"p-ex-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Imagine you have a local knowledge base about a game.\"},\"tunes\":{}},{\"id\":\"kb-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Knowledge base contains\",\"Example\"],[\"Weapons\",\"AKM uses 7.62 mm ammunition\"],[\"Healing items\",\"Med Kit restores health\"],[\"Attachments\",\"This attachment works with these weapons\"],[\"Map rules\",\"This zone behaves in this way\"]]},\"tunes\":{}},{\"id\":\"p-ex-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"You ask: “Which ammunition does the AKM use?”\"},\"tunes\":{}},{\"id\":\"p-ex-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG searches the knowledge base and finds the entry about the AKM. It gives that small piece of information to the LLM. The LLM then answers: “The AKM uses 7.62 mm ammunition.”\"},\"tunes\":{}},{\"id\":\"p-ex-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"The LLM did not need the entire database. RAG only brought the useful part.\"},\"tunes\":{}},{\"id\":\"h-state\",\"type\":\"header\",\"data\":{\"text\":\"Now the important part: RAG is not the current state\",\"level\":2},\"tunes\":{}},{\"id\":\"p-state-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"This is where many explanations become confusing.\"},\"tunes\":{}},{\"id\":\"p-state-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG usually gives the AI knowledge. A state system gives the AI facts about what is true right now.\"},\"tunes\":{}},{\"id\":\"knowledge-state\",\"type\":\"comparison\",\"data\":{\"title\":\"Knowledge vs current state\",\"layout\":\"table\",\"columns\":[{\"id\":\"knowledge\",\"label\":\"RAG \u002F knowledge\"},{\"id\":\"state\",\"label\":\"Current state\"}],\"rows\":[{\"id\":\"weapon\",\"label\":\"Weapon\",\"values\":[\"\",\"\"]},{\"id\":\"ammo\",\"label\":\"Ammunition\",\"values\":[\"\",\"\"]},{\"id\":\"health\",\"label\":\"Health\",\"values\":[\"\",\"\"]},{\"id\":\"enemy\",\"label\":\"Enemy\",\"values\":[\"\",\"\"]}]},\"tunes\":{}},{\"id\":\"dont-mix\",\"type\":\"callout\",\"data\":{\"variant\":\"warning\",\"title\":\"Do not mix these two\",\"body\":\"RAG answers: \u003Cstrong>What is generally true?\u003C\u002Fstrong>\u003Cbr>State answers: \u003Cstrong>What is true right now?\u003C\u002Fstrong>\"},\"tunes\":{}},{\"id\":\"h-state-db\",\"type\":\"header\",\"data\":{\"text\":\"What is a state database?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-statedb-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"A state database or state store is simply a place where the application keeps current facts.\"},\"tunes\":{}},{\"id\":\"p-statedb-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"In a game, the engine already knows things such as your health, position, inventory, ammunition, current mission, nearby objects and enemy status. An AI system can expose selected parts of that state to the model.\"},\"tunes\":{}},{\"id\":\"p-statedb-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"In a business application, the same idea could be an order database, a customer record, a project status or the current value of a sensor.\"},\"tunes\":{}},{\"id\":\"p-statedb-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"The state is created by the application itself as things happen. If you lose health, the game updates the health value. If you pick up ammunition, the inventory changes. If an order is paid, the business system changes the order status.\"},\"tunes\":{}},{\"id\":\"state-rule\",\"type\":\"callout\",\"data\":{\"variant\":\"info\",\"title\":\"Simple rule\",\"body\":\"The application creates and updates \u003Cstrong>state\u003C\u002Fstrong>. RAG searches \u003Cstrong>knowledge\u003C\u002Fstrong>. The LLM uses both to decide what to say or do.\"},\"tunes\":{}},{\"id\":\"h-together\",\"type\":\"header\",\"data\":{\"text\":\"How the three pieces work together\",\"level\":2},\"tunes\":{}},{\"id\":\"together-flow\",\"type\":\"processFlow\",\"data\":{\"title\":\"LLM + state + RAG\",\"orientation\":\"auto\",\"steps\":[{\"label\":\"1. Current state\",\"description\":\"The application tells the AI what is true now: health 41%, AKM equipped, 23 rounds.\"},{\"label\":\"2. RAG\",\"description\":\"The system retrieves useful knowledge: how the weapon works, which healing item is available, or a relevant rule.\"},{\"label\":\"3. LLM\",\"description\":\"The model receives the question, current state and retrieved knowledge.\"},{\"label\":\"4. Reasoning\",\"description\":\"The LLM combines those inputs and decides what answer or high-level action makes sense.\"},{\"label\":\"5. Application\",\"description\":\"If an action is required, the application or game engine executes it and updates the state again.\"}]},\"tunes\":{}},{\"id\":\"p-arch-intro\",\"type\":\"paragraph\",\"data\":{\"text\":\"So the basic architecture is:\"},\"tunes\":{}},{\"id\":\"simple-architecture\",\"type\":\"callout\",\"data\":{\"variant\":\"note\",\"title\":\"The simplest architecture\",\"body\":\"\u003Cstrong>State = what is true now\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = useful knowledge\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>LLM = understands, reasons and writes\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Application = performs the real action\u003C\u002Fstrong>\"},\"tunes\":{}},{\"id\":\"h-vector\",\"type\":\"header\",\"data\":{\"text\":\"Does RAG always use a vector database?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-vector-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"No.\"},\"tunes\":{}},{\"id\":\"p-vector-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"A vector database is a common way to build semantic search, but it is not the definition of RAG.\"},\"tunes\":{}},{\"id\":\"p-vector-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"The important part is retrieval: the system finds relevant external information and adds it to the LLM's context before the answer is generated.\"},\"tunes\":{}},{\"id\":\"p-vector-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"OpenAI's File Search, for example, can work with files stored in vector stores. Files are chunked into smaller pieces so the system can retrieve the parts that are relevant to a question. That is one implementation of the same basic idea.\"},\"tunes\":{}},{\"id\":\"h-embedding\",\"type\":\"header\",\"data\":{\"text\":\"What is an embedding, in plain English?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-emb-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"You do not need to understand embeddings to understand RAG.\"},\"tunes\":{}},{\"id\":\"p-emb-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"But the simple version is this: an embedding is a numerical representation of meaning. It helps a search system find text that is conceptually similar even when the words are not exactly the same.\"},\"tunes\":{}},{\"id\":\"p-emb-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"For example, a normal keyword search may look for the exact words “car repair.” Semantic search can also understand that “fix my vehicle” is about a similar topic.\"},\"tunes\":{}},{\"id\":\"p-emb-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"That makes embeddings useful for RAG, but RAG can also use keyword search, database queries or a hybrid of several methods.\"},\"tunes\":{}},{\"id\":\"h-memory\",\"type\":\"header\",\"data\":{\"text\":\"RAG is not memory either\",\"level\":2},\"tunes\":{}},{\"id\":\"p-memory-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Memory is another concept that is often mixed together with RAG.\"},\"tunes\":{}},{\"id\":\"p-memory-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"Memory is usually information the system keeps about previous interactions or previous events. RAG is the mechanism used to retrieve relevant knowledge when it is needed.\"},\"tunes\":{}},{\"id\":\"parts-table\",\"type\":\"table\",\"data\":{\"withHeadings\":true,\"stretched\":false,\"content\":[[\"Part\",\"Simple meaning\"],[\"LLM\",\"The part that understands and generates language\"],[\"RAG\",\"The part that looks up relevant knowledge before the answer\"],[\"Knowledge base\",\"The information RAG can search\"],[\"State\",\"What is true right now in the application or world\"],[\"Memory\",\"Information kept from previous interactions or events\"],[\"Tool \u002F action\",\"Something the AI is allowed to call or ask the application to do\"],[\"Context\",\"The information currently placed in front of the LLM for this request\"]]},\"tunes\":{}},{\"id\":\"h-pubg\",\"type\":\"header\",\"data\":{\"text\":\"A real game example: PUBG Ally\",\"level\":2},\"tunes\":{}},{\"id\":\"p-pubg-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"PUBG Ally is a useful example because it makes the difference visible.\"},\"tunes\":{}},{\"id\":\"p-pubg-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"KRAFTON describes live match state as a separate source of truth. The game exposes current facts through observation tools: current weapon, ammunition, health, safe-zone status, nearby items and combat situation.\"},\"tunes\":{}},{\"id\":\"p-pubg-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"Knowledge lookup is a different job. The system can use curated knowledge about weapons, attachments, items and rules. NVIDIA's ACE Game Agent SDK also exposes a separate RAG API for retrieving knowledge from developer-built databases.\"},\"tunes\":{}},{\"id\":\"p-pubg-4\",\"type\":\"paragraph\",\"data\":{\"text\":\"That gives us the clean separation: the game engine says what is happening now, retrieval provides relevant knowledge, and the language model decides what the information means.\"},\"tunes\":{}},{\"id\":\"ref-pubg\",\"type\":\"referralArticle\",\"data\":{\"url\":\"https:\u002F\u002Ffigure.rocks\u002Fblog\u002Fpubg-ally-shows-why-ai-teammates-need-two-brains-fast-reflexes-and-slow-reasoning\",\"title\":\"PUBG Ally Shows Why AI Teammates Need Two Brains: Fast Reflexes and Slow Reasoning\",\"excerpt\":\"A practical game example showing how live state, language reasoning and deterministic game-side control can work together.\",\"ctaLabel\":\"Read the PUBG Ally architecture article\"},\"tunes\":{}},{\"id\":\"h-complete\",\"type\":\"header\",\"data\":{\"text\":\"One complete example\",\"level\":2},\"tunes\":{}},{\"id\":\"p-complete-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Imagine you tell an AI teammate: “I am low on health. Should we attack?”\"},\"tunes\":{}},{\"id\":\"complete-flow\",\"type\":\"processFlow\",\"data\":{\"title\":\"What happens next\",\"orientation\":\"auto\",\"steps\":[{\"label\":\"State\",\"description\":\"The game reports: health 24%, one enemy nearby, two healing items available.\"},{\"label\":\"RAG\",\"description\":\"The knowledge system retrieves the relevant rules for the healing item and perhaps information about the current weapon or tactical mechanic.\"},{\"label\":\"LLM\",\"description\":\"The model combines your request, the current state and the retrieved knowledge.\"},{\"label\":\"Decision\",\"description\":\"It concludes that healing first is safer than attacking immediately.\"},{\"label\":\"Tool \u002F game engine\",\"description\":\"The agent requests a legal game action such as moving to cover or using the healing item.\"},{\"label\":\"New state\",\"description\":\"The game executes the action and reports the updated situation back to the agent.\"}]},\"tunes\":{}},{\"id\":\"p-complete-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG did not control the character. The state database did not reason. The LLM did not directly change the game. Each part had one job.\"},\"tunes\":{}},{\"id\":\"h-why\",\"type\":\"header\",\"data\":{\"text\":\"Why use RAG at all?\",\"level\":2},\"tunes\":{}},{\"id\":\"p-why-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"Because putting every document, rule and database record into every prompt would be slow, expensive and often confusing.\"},\"tunes\":{}},{\"id\":\"p-why-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG lets the system select only the information that is useful for the current question.\"},\"tunes\":{}},{\"id\":\"p-why-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"It also lets you update the knowledge base without retraining the entire language model. Change the document or database, rebuild or refresh the index when necessary, and the next retrieval can use the newer information.\"},\"tunes\":{}},{\"id\":\"h-not-guarantee\",\"type\":\"header\",\"data\":{\"text\":\"What RAG does not guarantee\",\"level\":2},\"tunes\":{}},{\"id\":\"p-not-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG can improve grounding, but it does not make an answer automatically correct.\"},\"tunes\":{}},{\"id\":\"p-not-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The retrieval step can find the wrong document. The correct document can be outdated. The LLM can misunderstand good evidence. Or the current state can have changed.\"},\"tunes\":{}},{\"id\":\"p-not-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"A reliable system therefore has to validate retrieval, state freshness and the model's final reasoning separately.\"},\"tunes\":{}},{\"id\":\"h-mental\",\"type\":\"header\",\"data\":{\"text\":\"The easiest mental model to remember\",\"level\":2},\"tunes\":{}},{\"id\":\"mental-table\",\"type\":\"comparison\",\"data\":{\"title\":\"Think of an AI system like a person at a desk\",\"layout\":\"table\",\"columns\":[{\"id\":\"analogy\",\"label\":\"Analogy\"},{\"id\":\"system\",\"label\":\"AI system\"}],\"rows\":[{\"id\":\"brain\",\"label\":\"Person thinking\",\"values\":[\"\",\"\"]},{\"id\":\"library\",\"label\":\"Finding a reference book\",\"values\":[\"\",\"\"]},{\"id\":\"books\",\"label\":\"Books on the shelf\",\"values\":[\"\",\"\"]},{\"id\":\"dashboard\",\"label\":\"Current dashboard or instrument panel\",\"values\":[\"\",\"\"]},{\"id\":\"notes\",\"label\":\"Notes from earlier meetings\",\"values\":[\"\",\"\"]},{\"id\":\"hands\",\"label\":\"Doing something in the real world\",\"values\":[\"\",\"\"]}]},\"tunes\":{}},{\"id\":\"remember\",\"type\":\"callout\",\"data\":{\"variant\":\"success\",\"title\":\"If you remember only this\",\"body\":\"\u003Cstrong>LLM = brain.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = librarian.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Knowledge base = library.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>State = what the dashboard says right now.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Tools = the hands that can actually do something.\u003C\u002Fstrong>\"},\"tunes\":{}},{\"id\":\"h-conclusion\",\"type\":\"header\",\"data\":{\"text\":\"Conclusion\",\"level\":2},\"tunes\":{}},{\"id\":\"p-conc-1\",\"type\":\"paragraph\",\"data\":{\"text\":\"RAG is much less mysterious once the parts are separated.\"},\"tunes\":{}},{\"id\":\"p-conc-2\",\"type\":\"paragraph\",\"data\":{\"text\":\"The LLM understands and generates language. The application maintains current state. The knowledge base stores information. RAG finds the useful part of that information and puts it into the LLM's context. Tools or the application perform real actions.\"},\"tunes\":{}},{\"id\":\"p-conc-3\",\"type\":\"paragraph\",\"data\":{\"text\":\"That is the basic architecture behind many modern AI assistants and agents.\"},\"tunes\":{}},{\"id\":\"h-faq\",\"type\":\"header\",\"data\":{\"text\":\"FAQ\",\"level\":2},\"tunes\":{}},{\"id\":\"faq\",\"type\":\"faq\",\"data\":{\"title\":\"RAG in plain English\",\"items\":[{\"id\":\"faq1\",\"question\":\"What is RAG in simple terms?\",\"answer\":\"RAG is a step where an AI searches a knowledge source for relevant information before the language model writes its answer.\"},{\"id\":\"faq2\",\"question\":\"Does RAG need the Internet?\",\"answer\":\"No. The knowledge base can be completely local on your computer or server.\"},{\"id\":\"faq3\",\"question\":\"Is RAG the same as a database?\",\"answer\":\"No. The database or files contain the information. RAG is the retrieval process that finds the useful part and gives it to the LLM.\"},{\"id\":\"faq4\",\"question\":\"Is RAG the same as memory?\",\"answer\":\"No. Memory usually stores previous interactions or events. RAG retrieves relevant knowledge when it is needed.\"},{\"id\":\"faq5\",\"question\":\"Is current application state part of RAG?\",\"answer\":\"Not necessarily. Current state is usually obtained directly from the application or a state store. RAG is better understood as retrieval from a knowledge source.\"},{\"id\":\"faq6\",\"question\":\"Does RAG make AI answers correct?\",\"answer\":\"No. It can provide better evidence, but retrieval can still be wrong or outdated and the LLM can still reason incorrectly.\"}]},\"tunes\":{}},{\"id\":\"h-glossary\",\"type\":\"header\",\"data\":{\"text\":\"Glossary\",\"level\":2},\"tunes\":{}},{\"id\":\"glossary\",\"type\":\"glossary\",\"data\":{\"title\":\"The basic terms\",\"entries\":[{\"term\":\"LLM\",\"definition\":\"A language model that understands and generates text and can reason over information placed in its context.\",\"anchor\":\"llm\"},{\"term\":\"RAG\",\"definition\":\"Retrieval-Augmented Generation: retrieving relevant external information and adding it to the model's context before generating an answer.\",\"anchor\":\"rag\"},{\"term\":\"Knowledge base\",\"definition\":\"The files, documents, records or other information that retrieval can search.\",\"anchor\":\"knowledge-base\"},{\"term\":\"State\",\"definition\":\"The current facts of an application, system or world at a particular moment.\",\"anchor\":\"state\"},{\"term\":\"Context\",\"definition\":\"The information currently supplied to the language model for one request or reasoning step.\",\"anchor\":\"context\"},{\"term\":\"Embedding\",\"definition\":\"A numerical representation of meaning that can help semantic search find conceptually similar information.\",\"anchor\":\"embedding\"}]},\"tunes\":{}},{\"id\":\"h-sources\",\"type\":\"header\",\"data\":{\"text\":\"Primary sources\",\"level\":2},\"tunes\":{}},{\"id\":\"src-openai-vector\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fapi-reference\u002Fvector-stores-files\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"OpenAI — Vector Store Files\",\"description\":\"Official documentation showing how files can be attached to vector stores, chunked and made available to file-search retrieval.\"}},\"tunes\":{}},{\"id\":\"src-openai-quickstart\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fplatform.openai.com\u002Fdocs\u002Fquickstart\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"OpenAI — Developer Quickstart\",\"description\":\"Official OpenAI documentation describing tools such as file search for giving models access to external information.\"}},\"tunes\":{}},{\"id\":\"src-nvidia-ace\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fdeveloper.nvidia.com\u002Face-for-games\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NVIDIA Developer — ACE for Games\",\"description\":\"Official NVIDIA documentation describing separate Agent, Chat and RAG APIs for connecting game characters to game state, contextual knowledge and model-driven actions.\"}},\"tunes\":{}},{\"id\":\"src-nvidia-pubg\",\"type\":\"linkTool\",\"data\":{\"link\":\"https:\u002F\u002Fdeveloper.nvidia.com\u002Fblog\u002Fhow-krafton-built-pubg-ally-a-co-playable-character-powered-by-nvidia-ace\u002F\",\"meta\":{\"image\":{\"url\":\"\"},\"title\":\"NVIDIA Developer — How KRAFTON Built PUBG Ally\",\"description\":\"Official technical explanation separating live match state from knowledge lookup and language-model reasoning.\"}},\"tunes\":{}}],\"version\":\"2.31.6\"}",{"time":926,"blocks":927,"version":1450},1790377494031,[928,932,937,941,945,949,953,957,961,966,970,974,978,982,987,991,1008,1012,1016,1020,1024,1043,1047,1051,1055,1059,1063,1067,1089,1094,1098,1102,1106,1110,1114,1118,1122,1140,1144,1149,1153,1157,1161,1165,1169,1173,1177,1181,1185,1189,1193,1197,1201,1227,1231,1235,1239,1243,1247,1253,1257,1261,1281,1285,1289,1293,1297,1301,1305,1309,1313,1317,1321,1349,1354,1358,1362,1366,1370,1374,1397,1401,1418,1422,1429,1436,1443],{"id":215,"data":929,"type":218,"tunes":931},{"text":930},"RAG sounds complicated because the name is complicated. The idea is not. RAG simply means: before the AI answers, it first looks up relevant information from a knowledge source and gives that information to the language model.",{},{"id":221,"data":933,"type":226,"tunes":936},{"body":934,"title":935,"variant":225},"\u003Cstrong>RAG is the step where an AI searches a knowledge base for useful information before the LLM writes the answer.\u003C\u002Fstrong>","RAG in one sentence",{},{"id":229,"data":938,"type":218,"tunes":940},{"text":939},"Think of an LLM as a smart person sitting at a desk. RAG is the librarian who brings the right page from the right book. The LLM then reads that page and answers you.",{},{"id":234,"data":942,"type":239,"tunes":944},{"title":943,"maxLevel":237,"minLevel":238},"Contents",{},{"id":242,"data":946,"type":42,"tunes":948},{"text":947,"level":238},"First: what does the LLM do?",{},{"id":247,"data":950,"type":218,"tunes":952},{"text":951},"The LLM is the part that understands language and produces language. It can read your question, understand instructions, compare information, explain something and write an answer.",{},{"id":252,"data":954,"type":218,"tunes":956},{"text":955},"But the LLM does not automatically know what is currently inside your company database, your game session, your private documents or a file you created five minutes ago.",{},{"id":257,"data":958,"type":218,"tunes":960},{"text":959},"It only knows what is already inside the model plus whatever information the application gives it in the current request.",{},{"id":262,"data":962,"type":226,"tunes":965},{"body":963,"title":964,"variant":266},"The LLM \u003Cstrong>thinks and writes\u003C\u002Fstrong>. It does not automatically own all of your current data.","Simple rule",{},{"id":269,"data":967,"type":42,"tunes":969},{"text":968,"level":238},"Then: what is the knowledge base?",{},{"id":274,"data":971,"type":218,"tunes":973},{"text":972},"A knowledge base is simply information the application can search.",{},{"id":279,"data":975,"type":218,"tunes":977},{"text":976},"It could contain PDFs, manuals, product documentation, support articles, contracts, game rules, weapon data, internal company documents, database records or other text.",{},{"id":284,"data":979,"type":218,"tunes":981},{"text":980},"The knowledge base can be local on your own machine. It can be on a server. It can be in a vector database. It can also be built from normal files. RAG does not mean Internet.",{},{"id":289,"data":983,"type":226,"tunes":986},{"body":984,"title":985,"variant":293},"\u003Cstrong>RAG does not require the Internet.\u003C\u002Fstrong> The information can be completely local.","Important",{},{"id":296,"data":988,"type":42,"tunes":990},{"text":989,"level":238},"So what does RAG actually do?",{},{"id":301,"data":992,"type":318,"tunes":1007},{"steps":993,"title":1006,"orientation":317},[994,997,1000,1003],{"label":995,"description":996},"1. You ask a question","For example: Which ammunition does this weapon use?",{"label":998,"description":999},"2. RAG searches the knowledge base","The system looks for the small pieces of information most relevant to your question.",{"label":1001,"description":1002},"3. RAG gives those pieces to the LLM","The LLM receives the question plus the retrieved information.",{"label":1004,"description":1005},"4. The LLM writes the answer","It uses the retrieved information as context for the response.","The whole RAG process",{},{"id":321,"data":1009,"type":218,"tunes":1011},{"text":1010},"That is RAG.",{},{"id":326,"data":1013,"type":218,"tunes":1015},{"text":1014},"The full name is Retrieval-Augmented Generation. Retrieval means finding the relevant information. Augmented means adding that information to the model's context. Generation means the LLM writes the final answer.",{},{"id":331,"data":1017,"type":42,"tunes":1019},{"text":1018,"level":238},"A very simple example",{},{"id":336,"data":1021,"type":218,"tunes":1023},{"text":1022},"Imagine you have a local knowledge base about a game.",{},{"id":341,"data":1025,"type":359,"tunes":1042},{"content":1026,"stretched":43,"withHeadings":14},[1027,1030,1033,1036,1039],[1028,1029],"Knowledge base contains","Example",[1031,1032],"Weapons","AKM uses 7.62 mm ammunition",[1034,1035],"Healing items","Med Kit restores health",[1037,1038],"Attachments","This attachment works with these weapons",[1040,1041],"Map rules","This zone behaves in this way",{},{"id":362,"data":1044,"type":218,"tunes":1046},{"text":1045},"You ask: “Which ammunition does the AKM use?”",{},{"id":367,"data":1048,"type":218,"tunes":1050},{"text":1049},"RAG searches the knowledge base and finds the entry about the AKM. It gives that small piece of information to the LLM. The LLM then answers: “The AKM uses 7.62 mm ammunition.”",{},{"id":372,"data":1052,"type":218,"tunes":1054},{"text":1053},"The LLM did not need the entire database. RAG only brought the useful part.",{},{"id":377,"data":1056,"type":42,"tunes":1058},{"text":1057,"level":238},"Now the important part: RAG is not the current state",{},{"id":382,"data":1060,"type":218,"tunes":1062},{"text":1061},"This is where many explanations become confusing.",{},{"id":387,"data":1064,"type":218,"tunes":1066},{"text":1065},"RAG usually gives the AI knowledge. A state system gives the AI facts about what is true right now.",{},{"id":392,"data":1068,"type":420,"tunes":1088},{"rows":1069,"title":1082,"layout":359,"columns":1083},[1070,1073,1076,1079],{"id":396,"label":1071,"values":1072},"Weapon",[399,399],{"id":401,"label":1074,"values":1075},"Ammunition",[399,399],{"id":405,"label":1077,"values":1078},"Health",[399,399],{"id":409,"label":1080,"values":1081},"Enemy",[399,399],"Knowledge vs current state",[1084,1086],{"id":415,"label":1085},"RAG \u002F knowledge",{"id":418,"label":1087},"Current state",{},{"id":423,"data":1090,"type":226,"tunes":1093},{"body":1091,"title":1092,"variant":427},"RAG answers: \u003Cstrong>What is generally true?\u003C\u002Fstrong>\u003Cbr>State answers: \u003Cstrong>What is true right now?\u003C\u002Fstrong>","Do not mix these two",{},{"id":430,"data":1095,"type":42,"tunes":1097},{"text":1096,"level":238},"What is a state database?",{},{"id":435,"data":1099,"type":218,"tunes":1101},{"text":1100},"A state database or state store is simply a place where the application keeps current facts.",{},{"id":440,"data":1103,"type":218,"tunes":1105},{"text":1104},"In a game, the engine already knows things such as your health, position, inventory, ammunition, current mission, nearby objects and enemy status. An AI system can expose selected parts of that state to the model.",{},{"id":445,"data":1107,"type":218,"tunes":1109},{"text":1108},"In a business application, the same idea could be an order database, a customer record, a project status or the current value of a sensor.",{},{"id":450,"data":1111,"type":218,"tunes":1113},{"text":1112},"The state is created by the application itself as things happen. If you lose health, the game updates the health value. If you pick up ammunition, the inventory changes. If an order is paid, the business system changes the order status.",{},{"id":455,"data":1115,"type":226,"tunes":1117},{"body":1116,"title":964,"variant":225},"The application creates and updates \u003Cstrong>state\u003C\u002Fstrong>. RAG searches \u003Cstrong>knowledge\u003C\u002Fstrong>. The LLM uses both to decide what to say or do.",{},{"id":460,"data":1119,"type":42,"tunes":1121},{"text":1120,"level":238},"How the three pieces work together",{},{"id":465,"data":1123,"type":318,"tunes":1139},{"steps":1124,"title":1138,"orientation":317},[1125,1128,1130,1132,1135],{"label":1126,"description":1127},"1. Current state","The application tells the AI what is true now: health 41%, AKM equipped, 23 rounds.",{"label":472,"description":1129},"The system retrieves useful knowledge: how the weapon works, which healing item is available, or a relevant rule.",{"label":475,"description":1131},"The model receives the question, current state and retrieved knowledge.",{"label":1133,"description":1134},"4. Reasoning","The LLM combines those inputs and decides what answer or high-level action makes sense.",{"label":1136,"description":1137},"5. Application","If an action is required, the application or game engine executes it and updates the state again.","LLM + state + RAG",{},{"id":486,"data":1141,"type":218,"tunes":1143},{"text":1142},"So the basic architecture is:",{},{"id":491,"data":1145,"type":226,"tunes":1148},{"body":1146,"title":1147,"variant":266},"\u003Cstrong>State = what is true now\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = useful knowledge\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>LLM = understands, reasons and writes\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Application = performs the real action\u003C\u002Fstrong>","The simplest architecture",{},{"id":497,"data":1150,"type":42,"tunes":1152},{"text":1151,"level":238},"Does RAG always use a vector database?",{},{"id":502,"data":1154,"type":218,"tunes":1156},{"text":1155},"No.",{},{"id":507,"data":1158,"type":218,"tunes":1160},{"text":1159},"A vector database is a common way to build semantic search, but it is not the definition of RAG.",{},{"id":512,"data":1162,"type":218,"tunes":1164},{"text":1163},"The important part is retrieval: the system finds relevant external information and adds it to the LLM's context before the answer is generated.",{},{"id":517,"data":1166,"type":218,"tunes":1168},{"text":1167},"OpenAI's File Search, for example, can work with files stored in vector stores. Files are chunked into smaller pieces so the system can retrieve the parts that are relevant to a question. That is one implementation of the same basic idea.",{},{"id":522,"data":1170,"type":42,"tunes":1172},{"text":1171,"level":238},"What is an embedding, in plain English?",{},{"id":527,"data":1174,"type":218,"tunes":1176},{"text":1175},"You do not need to understand embeddings to understand RAG.",{},{"id":532,"data":1178,"type":218,"tunes":1180},{"text":1179},"But the simple version is this: an embedding is a numerical representation of meaning. It helps a search system find text that is conceptually similar even when the words are not exactly the same.",{},{"id":537,"data":1182,"type":218,"tunes":1184},{"text":1183},"For example, a normal keyword search may look for the exact words “car repair.” Semantic search can also understand that “fix my vehicle” is about a similar topic.",{},{"id":542,"data":1186,"type":218,"tunes":1188},{"text":1187},"That makes embeddings useful for RAG, but RAG can also use keyword search, database queries or a hybrid of several methods.",{},{"id":547,"data":1190,"type":42,"tunes":1192},{"text":1191,"level":238},"RAG is not memory either",{},{"id":552,"data":1194,"type":218,"tunes":1196},{"text":1195},"Memory is another concept that is often mixed together with RAG.",{},{"id":557,"data":1198,"type":218,"tunes":1200},{"text":1199},"Memory is usually information the system keeps about previous interactions or previous events. RAG is the mechanism used to retrieve relevant knowledge when it is needed.",{},{"id":562,"data":1202,"type":359,"tunes":1226},{"content":1203,"stretched":43,"withHeadings":14},[1204,1207,1209,1211,1214,1217,1220,1223],[1205,1206],"Part","Simple meaning",[569,1208],"The part that understands and generates language",[572,1210],"The part that looks up relevant knowledge before the answer",[1212,1213],"Knowledge base","The information RAG can search",[1215,1216],"State","What is true right now in the application or world",[1218,1219],"Memory","Information kept from previous interactions or events",[1221,1222],"Tool \u002F action","Something the AI is allowed to call or ask the application to do",[1224,1225],"Context","The information currently placed in front of the LLM for this request",{},{"id":591,"data":1228,"type":42,"tunes":1230},{"text":1229,"level":238},"A real game example: PUBG Ally",{},{"id":596,"data":1232,"type":218,"tunes":1234},{"text":1233},"PUBG Ally is a useful example because it makes the difference visible.",{},{"id":601,"data":1236,"type":218,"tunes":1238},{"text":1237},"KRAFTON describes live match state as a separate source of truth. The game exposes current facts through observation tools: current weapon, ammunition, health, safe-zone status, nearby items and combat situation.",{},{"id":606,"data":1240,"type":218,"tunes":1242},{"text":1241},"Knowledge lookup is a different job. The system can use curated knowledge about weapons, attachments, items and rules. NVIDIA's ACE Game Agent SDK also exposes a separate RAG API for retrieving knowledge from developer-built databases.",{},{"id":611,"data":1244,"type":218,"tunes":1246},{"text":1245},"That gives us the clean separation: the game engine says what is happening now, retrieval provides relevant knowledge, and the language model decides what the information means.",{},{"id":616,"data":1248,"type":622,"tunes":1252},{"url":618,"title":1249,"excerpt":1250,"ctaLabel":1251},"PUBG Ally Shows Why AI Teammates Need Two Brains: Fast Reflexes and Slow Reasoning","A practical game example showing how live state, language reasoning and deterministic game-side control can work together.","Read the PUBG Ally architecture article",{},{"id":625,"data":1254,"type":42,"tunes":1256},{"text":1255,"level":238},"One complete example",{},{"id":630,"data":1258,"type":218,"tunes":1260},{"text":1259},"Imagine you tell an AI teammate: “I am low on health. Should we attack?”",{},{"id":635,"data":1262,"type":318,"tunes":1280},{"steps":1263,"title":1279,"orientation":317},[1264,1266,1268,1270,1273,1276],{"label":1215,"description":1265},"The game reports: health 24%, one enemy nearby, two healing items available.",{"label":572,"description":1267},"The knowledge system retrieves the relevant rules for the healing item and perhaps information about the current weapon or tactical mechanic.",{"label":569,"description":1269},"The model combines your request, the current state and the retrieved knowledge.",{"label":1271,"description":1272},"Decision","It concludes that healing first is safer than attacking immediately.",{"label":1274,"description":1275},"Tool \u002F game engine","The agent requests a legal game action such as moving to cover or using the healing item.",{"label":1277,"description":1278},"New state","The game executes the action and reports the updated situation back to the agent.","What happens next",{},{"id":656,"data":1282,"type":218,"tunes":1284},{"text":1283},"RAG did not control the character. The state database did not reason. The LLM did not directly change the game. Each part had one job.",{},{"id":661,"data":1286,"type":42,"tunes":1288},{"text":1287,"level":238},"Why use RAG at all?",{},{"id":666,"data":1290,"type":218,"tunes":1292},{"text":1291},"Because putting every document, rule and database record into every prompt would be slow, expensive and often confusing.",{},{"id":671,"data":1294,"type":218,"tunes":1296},{"text":1295},"RAG lets the system select only the information that is useful for the current question.",{},{"id":676,"data":1298,"type":218,"tunes":1300},{"text":1299},"It also lets you update the knowledge base without retraining the entire language model. Change the document or database, rebuild or refresh the index when necessary, and the next retrieval can use the newer information.",{},{"id":681,"data":1302,"type":42,"tunes":1304},{"text":1303,"level":238},"What RAG does not guarantee",{},{"id":686,"data":1306,"type":218,"tunes":1308},{"text":1307},"RAG can improve grounding, but it does not make an answer automatically correct.",{},{"id":691,"data":1310,"type":218,"tunes":1312},{"text":1311},"The retrieval step can find the wrong document. The correct document can be outdated. The LLM can misunderstand good evidence. Or the current state can have changed.",{},{"id":696,"data":1314,"type":218,"tunes":1316},{"text":1315},"A reliable system therefore has to validate retrieval, state freshness and the model's final reasoning separately.",{},{"id":701,"data":1318,"type":42,"tunes":1320},{"text":1319,"level":238},"The easiest mental model to remember",{},{"id":706,"data":1322,"type":420,"tunes":1348},{"rows":1323,"title":1342,"layout":359,"columns":1343},[1324,1327,1330,1333,1336,1339],{"id":710,"label":1325,"values":1326},"Person thinking",[399,399],{"id":714,"label":1328,"values":1329},"Finding a reference book",[399,399],{"id":718,"label":1331,"values":1332},"Books on the shelf",[399,399],{"id":722,"label":1334,"values":1335},"Current dashboard or instrument panel",[399,399],{"id":726,"label":1337,"values":1338},"Notes from earlier meetings",[399,399],{"id":730,"label":1340,"values":1341},"Doing something in the real world",[399,399],"Think of an AI system like a person at a desk",[1344,1346],{"id":229,"label":1345},"Analogy",{"id":738,"label":1347},"AI system",{},{"id":742,"data":1350,"type":226,"tunes":1353},{"body":1351,"title":1352,"variant":293},"\u003Cstrong>LLM = brain.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>RAG = librarian.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Knowledge base = library.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>State = what the dashboard says right now.\u003C\u002Fstrong>\u003Cbr>\u003Cstrong>Tools = the hands that can actually do something.\u003C\u002Fstrong>","If you remember only this",{},{"id":748,"data":1355,"type":42,"tunes":1357},{"text":1356,"level":238},"Conclusion",{},{"id":753,"data":1359,"type":218,"tunes":1361},{"text":1360},"RAG is much less mysterious once the parts are separated.",{},{"id":758,"data":1363,"type":218,"tunes":1365},{"text":1364},"The LLM understands and generates language. The application maintains current state. The knowledge base stores information. RAG finds the useful part of that information and puts it into the LLM's context. Tools or the application perform real actions.",{},{"id":763,"data":1367,"type":218,"tunes":1369},{"text":1368},"That is the basic architecture behind many modern AI assistants and agents.",{},{"id":768,"data":1371,"type":42,"tunes":1373},{"text":1372,"level":238},"FAQ",{},{"id":773,"data":1375,"type":773,"tunes":1396},{"items":1376,"title":1395},[1377,1380,1383,1386,1389,1392],{"id":777,"answer":1378,"question":1379},"RAG is a step where an AI searches a knowledge source for relevant information before the language model writes its answer.","What is RAG in simple terms?",{"id":781,"answer":1381,"question":1382},"No. The knowledge base can be completely local on your computer or server.","Does RAG need the Internet?",{"id":785,"answer":1384,"question":1385},"No. The database or files contain the information. RAG is the retrieval process that finds the useful part and gives it to the LLM.","Is RAG the same as a database?",{"id":789,"answer":1387,"question":1388},"No. Memory usually stores previous interactions or events. RAG retrieves relevant knowledge when it is needed.","Is RAG the same as memory?",{"id":793,"answer":1390,"question":1391},"Not necessarily. Current state is usually obtained directly from the application or a state store. RAG is better understood as retrieval from a knowledge source.","Is current application state part of RAG?",{"id":797,"answer":1393,"question":1394},"No. It can provide better evidence, but retrieval can still be wrong or outdated and the LLM can still reason incorrectly.","Does RAG make AI answers correct?","RAG in plain English",{},{"id":803,"data":1398,"type":42,"tunes":1400},{"text":1399,"level":238},"Glossary",{},{"id":808,"data":1402,"type":808,"tunes":1417},{"title":1403,"entries":1404},"The basic terms",[1405,1407,1409,1411,1413,1415],{"term":569,"anchor":813,"definition":1406},"A language model that understands and generates text and can reason over information placed in its context.",{"term":572,"anchor":816,"definition":1408},"Retrieval-Augmented Generation: retrieving relevant external information and adding it to the model's context before generating an answer.",{"term":1212,"anchor":819,"definition":1410},"The files, documents, records or other information that retrieval can search.",{"term":1215,"anchor":418,"definition":1412},"The current facts of an application, system or world at a particular moment.",{"term":1224,"anchor":824,"definition":1414},"The information currently supplied to the language model for one request or reasoning step.",{"term":827,"anchor":828,"definition":1416},"A numerical representation of meaning that can help semantic search find conceptually similar information.",{},{"id":832,"data":1419,"type":42,"tunes":1421},{"text":1420,"level":238},"Primary sources",{},{"id":837,"data":1423,"type":844,"tunes":1428},{"link":839,"meta":1424},{"image":1425,"title":1426,"description":1427},{"url":399},"OpenAI — Vector Store Files","Official documentation showing how files can be attached to vector stores, chunked and made available to file-search retrieval.",{},{"id":847,"data":1430,"type":844,"tunes":1435},{"link":849,"meta":1431},{"image":1432,"title":1433,"description":1434},{"url":399},"OpenAI — Developer Quickstart","Official OpenAI documentation describing tools such as file search for giving models access to external information.",{},{"id":856,"data":1437,"type":844,"tunes":1442},{"link":858,"meta":1438},{"image":1439,"title":1440,"description":1441},{"url":399},"NVIDIA Developer — ACE for Games","Official NVIDIA documentation describing separate Agent, Chat and RAG APIs for connecting game characters to game state, contextual knowledge and model-driven actions.",{},{"id":865,"data":1444,"type":844,"tunes":1449},{"link":867,"meta":1445},{"image":1446,"title":1447,"description":1448},{"url":399},"NVIDIA Developer — How KRAFTON Built PUBG Ally","Official technical explanation separating live match state from knowledge lookup and language-model reasoning.",{},"2.31.6","RAG sounds complicated, but the idea is simple: before an AI answers, it first looks up useful information from a knowledge source and gives that information to the language model. This guide explains RAG, LLMs, state, memory and tools using one simple mental model.",{"lang":7,"title":208,"content":210,"contentJson":1453,"excerpt":874},{"time":212,"blocks":1454,"version":873},[1455,1458,1461,1464,1467,1470,1473,1476,1479,1482,1485,1488,1491,1494,1497,1500,1508,1511,1514,1517,1520,1529,1532,1535,1538,1541,1544,1547,1562,1565,1568,1571,1574,1577,1580,1583,1586,1595,1598,1601,1604,1607,1610,1613,1616,1619,1622,1625,1628,1631,1634,1637,1640,1652,1655,1658,1661,1664,1667,1670,1673,1676,1686,1689,1692,1695,1698,1701,1704,1707,1710,1713,1716,1735,1738,1741,1744,1747,1750,1753,1763,1766,1776,1779,1784,1789,1794],{"id":215,"data":1456,"type":218,"tunes":1457},{"text":217},{},{"id":221,"data":1459,"type":226,"tunes":1460},{"body":223,"title":224,"variant":225},{},{"id":229,"data":1462,"type":218,"tunes":1463},{"text":231},{},{"id":234,"data":1465,"type":239,"tunes":1466},{"title":236,"maxLevel":237,"minLevel":238},{},{"id":242,"data":1468,"type":42,"tunes":1469},{"text":244,"level":238},{},{"id":247,"data":1471,"type":218,"tunes":1472},{"text":249},{},{"id":252,"data":1474,"type":218,"tunes":1475},{"text":254},{},{"id":257,"data":1477,"type":218,"tunes":1478},{"text":259},{},{"id":262,"data":1480,"type":226,"tunes":1481},{"body":264,"title":265,"variant":266},{},{"id":269,"data":1483,"type":42,"tunes":1484},{"text":271,"level":238},{},{"id":274,"data":1486,"type":218,"tunes":1487},{"text":276},{},{"id":279,"data":1489,"type":218,"tunes":1490},{"text":281},{},{"id":284,"data":1492,"type":218,"tunes":1493},{"text":286},{},{"id":289,"data":1495,"type":226,"tunes":1496},{"body":291,"title":292,"variant":293},{},{"id":296,"data":1498,"type":42,"tunes":1499},{"text":298,"level":238},{},{"id":301,"data":1501,"type":318,"tunes":1507},{"steps":1502,"title":316,"orientation":317},[1503,1504,1505,1506],{"label":305,"description":306},{"label":308,"description":309},{"label":311,"description":312},{"label":314,"description":315},{},{"id":321,"data":1509,"type":218,"tunes":1510},{"text":323},{},{"id":326,"data":1512,"type":218,"tunes":1513},{"text":328},{},{"id":331,"data":1515,"type":42,"tunes":1516},{"text":333,"level":238},{},{"id":336,"data":1518,"type":218,"tunes":1519},{"text":338},{},{"id":341,"data":1521,"type":359,"tunes":1528},{"content":1522,"stretched":43,"withHeadings":14},[1523,1524,1525,1526,1527],[345,346],[348,349],[351,352],[354,355],[357,358],{},{"id":362,"data":1530,"type":218,"tunes":1531},{"text":364},{},{"id":367,"data":1533,"type":218,"tunes":1534},{"text":369},{},{"id":372,"data":1536,"type":218,"tunes":1537},{"text":374},{},{"id":377,"data":1539,"type":42,"tunes":1540},{"text":379,"level":238},{},{"id":382,"data":1542,"type":218,"tunes":1543},{"text":384},{},{"id":387,"data":1545,"type":218,"tunes":1546},{"text":389},{},{"id":392,"data":1548,"type":420,"tunes":1561},{"rows":1549,"title":412,"layout":359,"columns":1558},[1550,1552,1554,1556],{"id":396,"label":397,"values":1551},[399,399],{"id":401,"label":402,"values":1553},[399,399],{"id":405,"label":406,"values":1555},[399,399],{"id":409,"label":410,"values":1557},[399,399],[1559,1560],{"id":415,"label":416},{"id":418,"label":419},{},{"id":423,"data":1563,"type":226,"tunes":1564},{"body":425,"title":426,"variant":427},{},{"id":430,"data":1566,"type":42,"tunes":1567},{"text":432,"level":238},{},{"id":435,"data":1569,"type":218,"tunes":1570},{"text":437},{},{"id":440,"data":1572,"type":218,"tunes":1573},{"text":442},{},{"id":445,"data":1575,"type":218,"tunes":1576},{"text":447},{},{"id":450,"data":1578,"type":218,"tunes":1579},{"text":452},{},{"id":455,"data":1581,"type":226,"tunes":1582},{"body":457,"title":265,"variant":225},{},{"id":460,"data":1584,"type":42,"tunes":1585},{"text":462,"level":238},{},{"id":465,"data":1587,"type":318,"tunes":1594},{"steps":1588,"title":483,"orientation":317},[1589,1590,1591,1592,1593],{"label":469,"description":470},{"label":472,"description":473},{"label":475,"description":476},{"label":478,"description":479},{"label":481,"description":482},{},{"id":486,"data":1596,"type":218,"tunes":1597},{"text":488},{},{"id":491,"data":1599,"type":226,"tunes":1600},{"body":493,"title":494,"variant":266},{},{"id":497,"data":1602,"type":42,"tunes":1603},{"text":499,"level":238},{},{"id":502,"data":1605,"type":218,"tunes":1606},{"text":504},{},{"id":507,"data":1608,"type":218,"tunes":1609},{"text":509},{},{"id":512,"data":1611,"type":218,"tunes":1612},{"text":514},{},{"id":517,"data":1614,"type":218,"tunes":1615},{"text":519},{},{"id":522,"data":1617,"type":42,"tunes":1618},{"text":524,"level":238},{},{"id":527,"data":1620,"type":218,"tunes":1621},{"text":529},{},{"id":532,"data":1623,"type":218,"tunes":1624},{"text":534},{},{"id":537,"data":1626,"type":218,"tunes":1627},{"text":539},{},{"id":542,"data":1629,"type":218,"tunes":1630},{"text":544},{},{"id":547,"data":1632,"type":42,"tunes":1633},{"text":549,"level":238},{},{"id":552,"data":1635,"type":218,"tunes":1636},{"text":554},{},{"id":557,"data":1638,"type":218,"tunes":1639},{"text":559},{},{"id":562,"data":1641,"type":359,"tunes":1651},{"content":1642,"stretched":43,"withHeadings":14},[1643,1644,1645,1646,1647,1648,1649,1650],[566,567],[569,570],[572,573],[575,576],[578,579],[581,582],[584,585],[587,588],{},{"id":591,"data":1653,"type":42,"tunes":1654},{"text":593,"level":238},{},{"id":596,"data":1656,"type":218,"tunes":1657},{"text":598},{},{"id":601,"data":1659,"type":218,"tunes":1660},{"text":603},{},{"id":606,"data":1662,"type":218,"tunes":1663},{"text":608},{},{"id":611,"data":1665,"type":218,"tunes":1666},{"text":613},{},{"id":616,"data":1668,"type":622,"tunes":1669},{"url":618,"title":619,"excerpt":620,"ctaLabel":621},{},{"id":625,"data":1671,"type":42,"tunes":1672},{"text":627,"level":238},{},{"id":630,"data":1674,"type":218,"tunes":1675},{"text":632},{},{"id":635,"data":1677,"type":318,"tunes":1685},{"steps":1678,"title":653,"orientation":317},[1679,1680,1681,1682,1683,1684],{"label":578,"description":639},{"label":572,"description":641},{"label":569,"description":643},{"label":645,"description":646},{"label":648,"description":649},{"label":651,"description":652},{},{"id":656,"data":1687,"type":218,"tunes":1688},{"text":658},{},{"id":661,"data":1690,"type":42,"tunes":1691},{"text":663,"level":238},{},{"id":666,"data":1693,"type":218,"tunes":1694},{"text":668},{},{"id":671,"data":1696,"type":218,"tunes":1697},{"text":673},{},{"id":676,"data":1699,"type":218,"tunes":1700},{"text":678},{},{"id":681,"data":1702,"type":42,"tunes":1703},{"text":683,"level":238},{},{"id":686,"data":1705,"type":218,"tunes":1706},{"text":688},{},{"id":691,"data":1708,"type":218,"tunes":1709},{"text":693},{},{"id":696,"data":1711,"type":218,"tunes":1712},{"text":698},{},{"id":701,"data":1714,"type":42,"tunes":1715},{"text":703,"level":238},{},{"id":706,"data":1717,"type":420,"tunes":1734},{"rows":1718,"title":733,"layout":359,"columns":1731},[1719,1721,1723,1725,1727,1729],{"id":710,"label":711,"values":1720},[399,399],{"id":714,"label":715,"values":1722},[399,399],{"id":718,"label":719,"values":1724},[399,399],{"id":722,"label":723,"values":1726},[399,399],{"id":726,"label":727,"values":1728},[399,399],{"id":730,"label":731,"values":1730},[399,399],[1732,1733],{"id":229,"label":736},{"id":738,"label":739},{},{"id":742,"data":1736,"type":226,"tunes":1737},{"body":744,"title":745,"variant":293},{},{"id":748,"data":1739,"type":42,"tunes":1740},{"text":750,"level":238},{},{"id":753,"data":1742,"type":218,"tunes":1743},{"text":755},{},{"id":758,"data":1745,"type":218,"tunes":1746},{"text":760},{},{"id":763,"data":1748,"type":218,"tunes":1749},{"text":765},{},{"id":768,"data":1751,"type":42,"tunes":1752},{"text":770,"level":238},{},{"id":773,"data":1754,"type":773,"tunes":1762},{"items":1755,"title":800},[1756,1757,1758,1759,1760,1761],{"id":777,"answer":778,"question":779},{"id":781,"answer":782,"question":783},{"id":785,"answer":786,"question":787},{"id":789,"answer":790,"question":791},{"id":793,"answer":794,"question":795},{"id":797,"answer":798,"question":799},{},{"id":803,"data":1764,"type":42,"tunes":1765},{"text":805,"level":238},{},{"id":808,"data":1767,"type":808,"tunes":1775},{"title":810,"entries":1768},[1769,1770,1771,1772,1773,1774],{"term":569,"anchor":813,"definition":814},{"term":572,"anchor":816,"definition":817},{"term":575,"anchor":819,"definition":820},{"term":578,"anchor":418,"definition":822},{"term":587,"anchor":824,"definition":825},{"term":827,"anchor":828,"definition":829},{},{"id":832,"data":1777,"type":42,"tunes":1778},{"text":834,"level":238},{},{"id":837,"data":1780,"type":844,"tunes":1783},{"link":839,"meta":1781},{"image":1782,"title":842,"description":843},{"url":399},{},{"id":847,"data":1785,"type":844,"tunes":1788},{"link":849,"meta":1786},{"image":1787,"title":852,"description":853},{"url":399},{},{"id":856,"data":1790,"type":844,"tunes":1793},{"link":858,"meta":1791},{"image":1792,"title":861,"description":862},{"url":399},{},{"id":865,"data":1795,"type":844,"tunes":1798},{"link":867,"meta":1796},{"image":1797,"title":870,"description":871},{"url":399},{},"Post erfolgreich abgerufen",{"items":1801,"source":1884,"manualIds":1885,"manualMatchedIds":1886},[1802,1809,1816,1823,1830,1835,1842,1849,1856,1863,1870,1877],{"id":1803,"slug":1804,"title":1805,"excerpt":1806,"featuredImage":1807,"publishedAt":1808},"477","computer-use-agents-why-a-successful-demo-can-still-be-an-unreliable-system","Agenti za korišćenje računara: Zašto uspešan demo i dalje može biti nepouzdan sistem","Agenti za korišćenje računara sada mogu da završe impresivne radne tokove u pregledaču i na radnoj površini, ali jedno uspešno izvršavanje dokazuje sposobnost—ne pouzdanost. Ovaj članak pokazuje kako testirati ponovljivost, robusnost u odnosu na okruženje, kontrolu dugog horizonta, svest o stanju, verifikaciju ishoda i bezbedno upravljanje ciljevima.","\u002Fuploads\u002F2026\u002F09\u002Fcomputer-use-agents-why-a-successful-demo-can-still-be-an-unreliable-system-1790352854690-75qnrg.webp","2026-09-25T12:13:00.000Z",{"id":1810,"slug":1811,"title":1812,"excerpt":1813,"featuredImage":1814,"publishedAt":1815},"459","ollama-is-not-the-product-building-production-ready-open-llm-applications","Ollama nije proizvod: Izgradnja aplikacija spremnih za produkciju sa otvorenim LLM-ovima","Pokretanje lokalnog modela pomoću Ollama-e je jednostavno. Izgradnja Open-LLM aplikacije spremne za produkciju je teža: zahteva RAG, kontrolu pristupa, apstrakciju provajdera, evaluaciju, logovanje, disciplinu puštanja u rad i kontrolisani aplikativni sloj oko modela.","\u002Fuploads\u002F2026\u002F06\u002Follama-is-not-the-product-building-production-ready-open-llm-applications-1782679361640-h0usqf.webp","2026-06-28T16:39:00.000Z",{"id":1817,"slug":1818,"title":1819,"excerpt":1820,"featuredImage":1821,"publishedAt":1822},"469","rag-failed-but-which-layer-actually-failed-a-diagnostic-method","RAG nije uspeo — ali koji sloj je zapravo zakazao? Dijagnostička metoda","Kada je RAG odgovor pogrešan, kriviti pretragu ili model je previše neodređeno. Ova dijagnostička metoda izoluje pokrivenost izvora, konstrukciju upita, pretragu, rangiranje, sastavljanje konteksta, generisanje, pripisivanje dokaza i svežinu—tako da se stvarni kvar može reprodukovati i ispraviti.","\u002Fuploads\u002F2026\u002F09\u002Frag-failed-but-which-layer-actually-failed-a-diagnostic-method-1790350847177-pior4c.webp","2026-09-24T19:39:00.000Z",{"id":1824,"slug":1825,"title":1826,"excerpt":1827,"featuredImage":1828,"publishedAt":1829},"473","openai-agents-api-vs-agents-sdk-vs-responses-api-what-should-you-build-on-in-2026","OpenAI Agents API vs Agents SDK vs Responses API: Na čemu bi trebalo da gradite u 2026?","OpenAI-jev stek agenata promenio se u septembru 2026. Ovaj arhitekturni vodič razdvaja Agents API, Agents SDK, Responses API i Codex SDK prema vlasništvu nad izvršnim okruženjem—kako bi timovi mogli da izaberu pravu granicu kontrole umesto da porede nazive proizvoda.","\u002Fuploads\u002F2026\u002F09\u002Fopenai-agents-api-vs-agents-sdk-vs-responses-api-what-should-you-build-on-in-2026-1790351846714-zi7lus.webp","2026-09-25T11:56:00.000Z",{"id":1831,"slug":1832,"title":1832,"excerpt":10,"featuredImage":1833,"publishedAt":1834},"369","git-with-automatic-upload-and-synchronization-to-a-production-server","\u002Fuploads\u002F2024\u002F05\u002Fstep-by-step-guide-illustration-showing-the-process-of-setting-up-Git-with-auto-upload-and-synchronization-to-a-production-server-large.webp","2024-05-28T22:48:00.000Z",{"id":1836,"slug":1837,"title":1838,"excerpt":1839,"featuredImage":1840,"publishedAt":1841},"468","ai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context","Memorija AI agenta nije RAG: Kako razdvojiti memoriju, pronalaženje, stanje i kontekst","Memorija agenta, RAG, stanje i kontekst često se koriste kao da su međusobno zamenjivi. Oni to nisu. Ovaj praktični arhitektonski model razdvaja ova četiri sloja, pokazuje gde svaki pripada i objašnjava šta se kvari kada ih sistemi stope u jedno.","\u002Fuploads\u002F2026\u002F09\u002Fai-agent-memory-is-not-rag-how-to-separate-memory-retrieval-state-and-context-1790350560308-np0xy6.webp","2026-09-25T11:34:00.000Z",{"id":1843,"slug":1844,"title":1845,"excerpt":1846,"featuredImage":1847,"publishedAt":1848},"364","tipps-fuer-die-verbesserung-der-seo-suchmaschinenoptimierung","Ovladavanje SEO radnim tokom: Ključne strategije optimizacije za organski rast","Strukturiran SEO tok posla je ključan za održiv organski rast. Naučite deset osnovnih strategija, od istraživanja ključnih reči i tehničke optimizacije do kvaliteta sadržaja i analize performansi.","\u002Fuploads\u002F2026\u002F03\u002Ftipps-fuer-die-verbesserung-der-seo-suchmaschinenoptimierung-1774866098131-hwkzrg.webp","2024-01-26T06:35:00.000Z",{"id":1850,"slug":1851,"title":1852,"excerpt":1853,"featuredImage":1854,"publishedAt":1855},"476","mcp-vs-a2a-vs-ucp-vs-ap2-vs-a2ui-the-agent-protocol-stack-explained","MCP vs A2A vs UCP vs AP2 vs A2UI: Objašnjen stek agentskih protokola","MCP, A2A, UCP, AP2 i A2UI se često predstavljaju kao konkurentski standardi za agente. Oni uglavnom rešavaju različite probleme interoperabilnosti. Ovaj vodič mapira svaki protokol na granicu koju zapravo standardizuje—i pokazuje kako oni mogu da rade zajedno u jednom produkcionom sistemu.","\u002Fuploads\u002F2026\u002F09\u002Fmcp-vs-a2a-vs-ucp-vs-ap2-vs-a2ui-the-agent-protocol-stack-explained-1790352625869-2ezle0.webp","2026-09-25T12:09:00.000Z",{"id":1857,"slug":1858,"title":1859,"excerpt":1860,"featuredImage":1861,"publishedAt":1862},"470","what-should-an-ai-agent-remember-forget-recompute-or-retrieve-again","Šta bi AI agent trebalo da zapamti, zaboravi, ponovo izračuna ili ponovo preuzme?","Dugotrajni agenti ne bi trebalo da pamte sve. Ovaj članak pruža praktičan model životnog ciklusa za odlučivanje o tome šta pripada trajnoj memoriji, šta bi trebalo ponovo preuzeti, šta je bezbednije ponovo izračunati i šta bi trebalo da istekne ili bude zamenjeno.","\u002Fuploads\u002F2026\u002F09\u002Fwhat-should-an-ai-agent-remember-forget-recompute-or-retrieve-again-1790351131087-iehz28.webp","2026-09-25T09:43:00.000Z",{"id":1864,"slug":1865,"title":1866,"excerpt":1867,"featuredImage":1868,"publishedAt":1869},"467","the-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers","Granica valjanosti odgovora: Nedostajući sloj između relevantnosti i pouzdanih AI odgovora","Izvor može biti relevantan, autoritativan i ipak pogrešan za pitanje koje se postavlja. Sloj koji nedostaje je primenljivost: uslovi pod kojima odgovor važi i promene koje ga primoravaju na preispitivanje. Ovaj članak predstavlja Granicu važenja odgovora kao obrazac za dizajn izvora za ljude, AI pretragu i RAG sisteme.","\u002Fuploads\u002F2026\u002F09\u002Fthe-answer-validity-boundary-the-missing-layer-between-relevance-and-reliable-ai-answers-1790272901306-1g5jly.webp","2026-09-24T11:59:00.000Z",{"id":1871,"slug":1872,"title":1873,"excerpt":1874,"featuredImage":1875,"publishedAt":1876},"466","the-gpu-is-not-the-product-future-proof-private-ai-architecture","GPU nije proizvod: Privatna AI arhitektura spremna za budućnost","Privatna AI infrastruktura ne bi trebalo da bude projektovana oko jednog GPU-a ili jednog modela. Otporniji pristup kombinuje brze GPU-ove za inferenciju, memorijski bogate AI sisteme, čvorove za fizički AI i opcione vodeće modele u oblaku iza sloja za rutiranje koji prepoznaje mogućnosti.","\u002Fuploads\u002F2026\u002F09\u002Fthe-gpu-is-not-the-product-future-proof-private-ai-architecture-1790140878812-8hsl39.webp","2026-09-23T01:19:00.000Z",{"id":1878,"slug":1879,"title":1880,"excerpt":1881,"featuredImage":1882,"publishedAt":1883},"460","ai-agent-reliability-why-the-final-answer-is-not-enough","Pouzdanost AI agenata: Zašto konačni odgovor nije dovoljan","Tačan rezultat ne dokazuje ispravno razmišljanje, bezbedno izvršavanje ili pouzdan sistem.","\u002Fuploads\u002F2026\u002F09\u002Fai-agent-reliability-why-the-final-answer-is-not-enough-1788955466306-pl0qhz.webp","2026-09-09T04:01:00.000Z","fallback",[],[]]