{"id":6531,"date":"2026-08-04T00:10:39","date_gmt":"2026-08-03T22:10:39","guid":{"rendered":"https:\/\/www.hostx.ro\/kb\/"},"modified":"2026-08-04T00:11:09","modified_gmt":"2026-08-03T22:11:09","password":"","slug":"cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf","status":"publish","type":"docs","link":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/","title":{"rendered":"Cum rulez Hermes cu llama.cpp \u0219i fi\u0219iere GGUF?"},"content":{"rendered":"<p><strong>llama.cpp<\/strong> \u00ee\u021bi permite s\u0103 rulezi modele precum Hermes eficient, inclusiv pe <strong>CPU<\/strong> sau pe GPU-uri modeste, folosind fi\u0219iere \u00een format <strong>GGUF<\/strong> (cuantizate). Este o op\u021biune excelent\u0103 c\u00e2nd nu ai un GPU puternic.<\/p>\n<h2>Ce este GGUF<\/h2>\n<p>GGUF este formatul de fi\u0219ier folosit de llama.cpp pentru modele cuantizate. Vei g\u0103si variante Hermes \u00een GGUF pe Hugging Face, cu niveluri diferite de cuantizare (Q4_K_M, Q5_K_M, Q8 etc.).<\/p>\n<h2>1. Ob\u021bine llama.cpp<\/h2>\n<pre><code>git clone https:\/\/github.com\/ggerganov\/llama.cpp\r\ncd llama.cpp\r\ncmake -B build &amp;&amp; cmake --build build --config Release<\/code><\/pre>\n<h2>2. Descarc\u0103 un model Hermes \u00een GGUF<\/h2>\n<p>Descarc\u0103 fi\u0219ierul GGUF dorit (de exemplu o variant\u0103 8B, cuantizare Q4_K_M) de pe Hugging Face \u00een folderul <code>models<\/code>.<\/p>\n<h2>3. Ruleaz\u0103 modelul<\/h2>\n<pre><code>.\/build\/bin\/llama-cli -m models\/hermes-3-8b-Q4_K_M.gguf \\\r\n  -p \"Salut! Cine e\u0219ti?\"<\/code><\/pre>\n<h2>4. Porne\u0219te un server compatibil OpenAI<\/h2>\n<pre><code>.\/build\/bin\/llama-server -m models\/hermes-3-8b-Q4_K_M.gguf --port 8080<\/code><\/pre>\n<p>Astfel ob\u021bii un endpoint local pe care \u00eel po\u021bi folosi \u00een alte aplica\u021bii.<\/p>\n<h2>Alegerea cuantiz\u0103rii<\/h2>\n<ul>\n<li><strong>Q4_K_M<\/strong> \u2014 cel mai mic consum de memorie, calitate bun\u0103 (recomandat pentru \u00eenceput).<\/li>\n<li><strong>Q5 \/ Q8<\/strong> \u2014 calitate mai mare, consum de memorie mai mare.<\/li>\n<\/ul>\n<div class=\"note\">\n<p><strong>Sfat:<\/strong> Ollama folose\u0219te tot GGUF \u201e\u00een culise&#8221;. Dac\u0103 vrei simplitate, folose\u0219te Ollama; dac\u0103 vrei control fin asupra parametrilor, folose\u0219te llama.cpp direct.<\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>llama.cpp \u00ee\u021bi permite s\u0103 rulezi modele precum Hermes eficient, inclusiv pe CPU sau pe GPU-uri modeste, folosind fi\u0219iere \u00een format GGUF (cuantizate). Este o op\u021biune excelent\u0103 c\u00e2nd nu ai un GPU puternic. Ce este GGUF GGUF este formatul de fi\u0219ier folosit de llama.cpp pentru modele cuantizate. Vei g\u0103si variante Hermes \u00een GGUF pe Hugging Face, &#8230; <a title=\"Cum rulez Hermes cu llama.cpp \u0219i fi\u0219iere GGUF?\" class=\"read-more\" href=\"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/\" aria-label=\"Cite\u0219te mai multe despre Cum rulez Hermes cu llama.cpp \u0219i fi\u0219iere GGUF?\">Cite\u0219te mai mult<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","template":"","meta":{"inline_featured_image":false,"whmcs_kb_id":0,"footnotes":""},"doc_category":[499],"doc_tag":[262,437,227,508,243,421,304,300],"class_list":["post-6531","docs","type-docs","status-publish","hentry","doc_category-hermes-nous","doc_tag-api","doc_tag-cli","doc_tag-configurare","doc_tag-hermes","doc_tag-linia-de-comanda","doc_tag-node-js","doc_tag-server","doc_tag-terminal"],"year_month":"2026-11","word_count":231,"total_views":"24","reactions":{"happy":"0","normal":"0","sad":"0"},"author_info":{"name":"benyhill","author_nicename":"benyhill","author_url":"https:\/\/www.hostx.ro\/kb\/author\/benyhill\/"},"doc_category_info":[{"term_name":"Hermes (Nous Research)","term_url":"https:\/\/www.hostx.ro\/kb\/docs-category\/hermes-nous\/"}],"doc_tag_info":[{"term_name":"API","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/api\/"},{"term_name":"CLI","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/cli\/"},{"term_name":"configurare","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/configurare\/"},{"term_name":"Hermes","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/hermes\/"},{"term_name":"linia de comand\u0103","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/linia-de-comanda\/"},{"term_name":"Node.js","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/node-js\/"},{"term_name":"server","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/server\/"},{"term_name":"terminal","term_url":"https:\/\/www.hostx.ro\/kb\/docs-tag\/terminal\/"}],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX<\/title>\n<meta name=\"description\" content=\"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/\" \/>\n<meta property=\"og:locale\" content=\"ro_RO\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX\" \/>\n<meta property=\"og:description\" content=\"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/\" \/>\n<meta property=\"og:site_name\" content=\"Centrul de suport HostX\" \/>\n<meta property=\"article:modified_time\" content=\"2026-08-03T22:11:09+00:00\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Timp estimat pentru citire\" \/>\n\t<meta name=\"twitter:data1\" content=\"1 minut\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\\\/\",\"url\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\\\/\",\"name\":\"Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/#website\"},\"datePublished\":\"2026-08-03T22:10:39+00:00\",\"dateModified\":\"2026-08-03T22:11:09+00:00\",\"description\":\"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\\\/#breadcrumb\"},\"inLanguage\":\"ro-RO\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\\\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Prima pagin\u0103\",\"item\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"InfoCentru\",\"item\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/docs\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Cum rulez Hermes cu llama.cpp \u0219i fi\u0219iere GGUF?\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/#website\",\"url\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/\",\"name\":\"Centrul de suport HostX\",\"description\":\"tutoriale gazduire web, domenii web \u0219i servere\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.hostx.ro\\\/kb\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ro-RO\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX","description":"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/","og_locale":"ro_RO","og_type":"article","og_title":"Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX","og_description":"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.","og_url":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/","og_site_name":"Centrul de suport HostX","article_modified_time":"2026-08-03T22:11:09+00:00","twitter_card":"summary_large_image","twitter_misc":{"Timp estimat pentru citire":"1 minut"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/","url":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/","name":"Rulare Hermes cu llama.cpp \u0219i GGUF - Centrul de suport HostX","isPartOf":{"@id":"https:\/\/www.hostx.ro\/kb\/#website"},"datePublished":"2026-08-03T22:10:39+00:00","dateModified":"2026-08-03T22:11:09+00:00","description":"Ghid pentru rularea modelului Hermes cu llama.cpp \u0219i fi\u0219iere GGUF. Instalare, desc\u0103rcare model \u0219i configurare server OpenAI local.","breadcrumb":{"@id":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/#breadcrumb"},"inLanguage":"ro-RO","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.hostx.ro\/kb\/cum-rulez-hermes-cu-llama-cpp-si-fisiere-gguf\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Prima pagin\u0103","item":"https:\/\/www.hostx.ro\/kb\/"},{"@type":"ListItem","position":2,"name":"InfoCentru","item":"https:\/\/www.hostx.ro\/kb\/docs\/"},{"@type":"ListItem","position":3,"name":"Cum rulez Hermes cu llama.cpp \u0219i fi\u0219iere GGUF?"}]},{"@type":"WebSite","@id":"https:\/\/www.hostx.ro\/kb\/#website","url":"https:\/\/www.hostx.ro\/kb\/","name":"Centrul de suport HostX","description":"tutoriale gazduire web, domenii web \u0219i servere","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.hostx.ro\/kb\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ro-RO"}]}},"knowledge_base_info":[],"knowledge_base_slug":[],"_links":{"self":[{"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/docs\/6531","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/docs"}],"about":[{"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/types\/docs"}],"author":[{"embeddable":true,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/comments?post=6531"}],"version-history":[{"count":1,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/docs\/6531\/revisions"}],"predecessor-version":[{"id":6631,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/docs\/6531\/revisions\/6631"}],"wp:attachment":[{"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/media?parent=6531"}],"wp:term":[{"taxonomy":"doc_category","embeddable":true,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/doc_category?post=6531"},{"taxonomy":"doc_tag","embeddable":true,"href":"https:\/\/www.hostx.ro\/kb\/wp-json\/wp\/v2\/doc_tag?post=6531"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}