{"id":182824,"date":"2026-08-11T04:54:15","date_gmt":"2026-08-11T02:54:15","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3"},"modified":"2026-08-12T18:21:41","modified_gmt":"2026-08-12T16:21:41","slug":"yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","status":"publish","type":"post","link":"https:\/\/prohoster.info\/hu\/blog\/news\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","title":{"rendered":"A 28,9 milli\u00f3 param\u00e9terrel rendelkez\u0151 nyelvi modell m\u0171k\u00f6d\u00e9sbe l\u00e9pett az ESP32-S3-on","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

<\/i><\/a>
\nA fejleszt\u0151, Vjacseszlav Serbov, aki slvDev \u00e1ln\u00e9ven szerepel, bemutatta<\/a> teljesen lok\u00e1lis sz\u00f6veggener\u00e1l\u00e1st v\u00e9gez az ESP32-S3 mikrovez\u00e9rl\u0151n. \u00c1ltala l\u00e9trehozott nyelvi modell 28,9 milli\u00f3 param\u00e9tert tartalmaz, \u00e9s m\u00e1sodpercenk\u00e9nt k\u00f6r\u00fclbel\u00fcl 9,9 tokent ad ki, an\u00e9lk\u00fcl, hogy szerverhez vagy m\u00e1s k\u00fcls\u0151 sz\u00e1m\u00edt\u00e1si er\u0151forr\u00e1shoz fordulna. A gener\u00e1lt r\u00f6vid t\u00f6rt\u00e9netek egy csatlakoztatott OLED-k\u00e9perny\u0151n jelennek meg.<\/p>\n

 <\/p>\n

Projekt esp32-ai<\/a> az ESP32-S3 N16R8 modulon lett tesztelve, amely 512 KB be\u00e9p\u00edtett SRAM-mal, 8 MB PSRAM-mal \u00e9s 16 MB flash mem\u00f3ri\u00e1val rendelkezik. A n\u00e9gy bitre kvant\u00e1lt modell 14,9 MB-ot foglal el. A t\u00e1rol\u00f3ban k\u00f6zz\u00e9 lett t\u00e9ve a C nyelv\u0171 futtat\u00e1si k\u00f6r forr\u00e1sk\u00f3dja, a modell tan\u00edt\u00e1s\u00e1hoz \u00e9s kvant\u00e1l\u00e1s\u00e1hoz sz\u00fcks\u00e9ges eszk\u00f6z\u00f6k, firmware, tesztek \u00e9s a modulra val\u00f3 felt\u00f6lt\u00e9s szkriptjei.<\/p>\n

 <\/p>\n

A megadott 28,9 milli\u00f3 param\u00e9tert nem szabad k\u00f6zvetlen\u00fcl \u00f6sszehasonl\u00edtani a hagyom\u00e1nyos asztali LLM param\u00e9tereivel. A fejleszt\u0151 r\u00e9szletes jelent\u00e9se szerint,<\/a>, a modell egy s\u0171r\u0171 sz\u00e1m\u00edt\u00e1si magb\u00f3l \u00e1ll, k\u00f6r\u00fclbel\u00fcl 559 ezer param\u00e9terrel, egy kimeneti r\u00e9tegb\u0151l 3,1 milli\u00f3 param\u00e9terrel \u00e9s egy r\u00e9tegenk\u00e9nti vektor\u00e1br\u00e1zol\u00e1si t\u00e1bl\u00e1zatb\u00f3l k\u00f6r\u00fclbel\u00fcl 25 milli\u00f3 param\u00e9terrel. Ez ut\u00f3bbi biztos\u00edtja a modell form\u00e1lis m\u00e9ret\u00e9nek jelent\u0151s r\u00e9sz\u00e9t, de nem ker\u00fcl teljes eg\u00e9sz\u00e9ben feldolgoz\u00e1sra minden token gener\u00e1l\u00e1sakor.<\/p>\n

 <\/p>\n

A modell elhelyez\u00e9s\u00e9hez h\u00e1rom mem\u00f3ria szintet haszn\u00e1lnak. A leggyakrabban haszn\u00e1lt adatok a bels\u0151 SRAM-ban tal\u00e1lhat\u00f3k, a kimeneti r\u00e9teg, KV-cache \u00e9s ideiglenes pufferek a PSRAM-ban, m\u00edg a 25 milli\u00f3 param\u00e9ter\u0171 t\u00e1bl\u00e1zat flash mem\u00f3ri\u00e1ban marad \u00e9s az adresst\u00e9rbe t\u00e9rk\u00e9pezve \u00e9rhet\u0151 el. Minden tokenhez ebb\u0151l k\u00f6r\u00fclbel\u00fcl hat sor olvashat\u00f3, \u00f6sszesen k\u00f6r\u00fclbel\u00fcl 450 byte m\u00e9retben.<\/p>\n

 <\/p>\n

Ez a k\u00e9sz\u00fcl\u00e9k a Per-Layer Embeddings technol\u00f3gi\u00e1n alapul, amit a Google a Gemma 3n-ben alkalmazott. A Google dokument\u00e1ci\u00f3ja szerint,<\/a>, a PLE-param\u00e9terek a modell munkamem\u00f3ri\u00e1j\u00e1n k\u00edv\u00fcl t\u00e1rolhat\u00f3k, \u00e9s a kimeneti folyamat sor\u00e1n hozz\u00e1adhat\u00f3k, ahogy az egyes r\u00e9tegek feldolgoz\u00e1sa zajlik. Az esp32-ai ezt a megk\u00f6zel\u00edt\u00e9st \u00e1t\u00fcltette a mikrovez\u00e9rl\u0151 mem\u00f3riahierarchi\u00e1j\u00e1ra, ahol a gyors SRAM k\u00fcl\u00f6n\u00f6sen korl\u00e1tozott.<\/p>\n

 <\/p>\n

Az els\u0151 teljesen helyes, C nyelven \u00edrt futtat\u00e1si k\u00f6rnyezet csak 0,57 tokent adott m\u00e1sodpercenk\u00e9nt. Miut\u00e1n a kimeneti r\u00e9teget PSRAM-ra helyezt\u00e9k, \u00e1tt\u00e9rtek a nyolc bit\u00e9rintkez\u00e9ses aktiv\u00e1ci\u00f3kra, elosztott\u00e1k a sz\u00e1m\u00edt\u00e1sokat a k\u00e9t Xtensa LX7 mag k\u00f6z\u00f6tt, \u00e9s m\u00e1s optimaliz\u00e1l\u00e1sokat v\u00e9geztek, a k\u00e9sleltet\u00e9s 94,9 ms-ra cs\u00f6kkent a modell l\u00e9p\u00e9s\u00e9n\u00e9l. A v\u00e9gs\u0151 m\u00e9rt sebess\u00e9g 9,88 token m\u00e1sodpercenk\u00e9nt a teljes gener\u00e1l\u00e1si folyamat figyelembev\u00e9tel\u00e9vel. A f\u0151 korl\u00e1t most a PSRAM s\u00e1vsz\u00e9less\u00e9ge lett a kimeneti r\u00e9teg olvas\u00e1s\u00e1n\u00e1l.<\/p>\n

 <\/p>\n

A modell egy szintetikus adat\u00e1llom\u00e1nyon lett betan\u00edtva TinyStories<\/a>, amely egyszer\u0171 angol nyelv\u0171 t\u00f6rt\u00e9netekb\u0151l \u00e1ll, korl\u00e1tozott sz\u00f3kincsel. Ez\u00e9rt k\u00e9pes folytatni a mondatokat \u00e9s kicsi, \u00f6sszef\u00fcgg\u0151 t\u00f6rt\u00e9neteket alkotni, de nem alkalmas k\u00e9rd\u00e9sekre v\u00e1laszol\u00e1sra, utas\u00edt\u00e1sok v\u00e9grehajt\u00e1s\u00e1ra, programoz\u00e1sra vagy t\u00e9nyszer\u0171 tud\u00e1s el\u0151\u00e1ll\u00edt\u00e1s\u00e1ra. A fejleszt\u0151 ezt a munk\u00e1t els\u0151sorban a modell hat\u00e9kony mem\u00f3riak\u00e9sz\u00edt\u00e9s\u00e9nek bemutat\u00e1sak\u00e9nt tekinti, nem mint k\u00e9szen \u00e1ll\u00f3, \u00f6n\u00e1ll\u00f3 cseveg\u0151robotot.<\/p>\n

 <\/p>\n

A legfrissebb dokument\u00e1ci\u00f3ban a szerz\u0151 k\u00fcl\u00f6n hangs\u00falyozza, hogy az esp32-ai \u00f6n\u00e1ll\u00f3 fejleszt\u00e9s. llama2.c<\/a> Andrej Karpaty projektje \u00e9s a 260 ezer param\u00e9teres modell kor\u00e1bbi ind\u00edt\u00e1sa ESP32-S3-on csak ir\u00e1nyad\u00f3 p\u00e9ld\u00e1k \u00e9s el\u0151z\u0151 munk\u00e1k eml\u00edt\u00e9se: a k\u00f3d, a pontoz\u00e1sok \u00e9s a m\u00f3dszerek k\u00f6zvetlen\u00fcl nem lettek \u00e1tv\u00e9ve t\u0151l\u00fck. Az esp32-ai forr\u00e1sk\u00f3dja MIT licenc alatt terjesztik.<\/a>.<\/p>\n

 <\/p>\n

Forr\u00e1s: linux.org.ru<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u0420\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0447\u0438\u043a \u0412\u044f\u0447\u0435\u0441\u043b\u0430\u0432 \u0421\u0435\u0440\u0431\u043e\u0432, \u0432\u044b\u0441\u0442\u0443\u043f\u0430\u044e\u0449\u0438\u0439 \u043f\u043e\u0434 \u043f\u0441\u0435\u0432\u0434\u043e\u043d\u0438\u043c\u043e\u043c slvDev, \u043f\u0440\u043e\u0434\u0435\u043c\u043e\u043d\u0441\u0442\u0440\u0438\u0440\u043e\u0432\u0430\u043b \u043f\u043e\u043b\u043d\u043e\u0441\u0442\u044c\u044e \u043b\u043e\u043a\u0430\u043b\u044c\u043d\u0443\u044e \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0438\u044e \u0442\u0435\u043a\u0441\u0442\u0430 \u043d\u0430 \u043c\u0438\u043a\u0440\u043e\u043a\u043e\u043d\u0442\u0440\u043e\u043b\u043b\u0435\u0440\u0435 ESP32-S3. \u0421\u043e\u0437\u0434\u0430\u043d\u043d\u0430\u044f \u0438\u043c \u044f\u0437\u044b\u043a\u043e\u0432\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u0441\u043e\u0434\u0435\u0440\u0436\u0438\u0442 28,9 \u043c\u043b\u043d \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u043e\u0432 \u0438 \u0432\u044b\u0434\u0430\u0451\u0442 \u043e\u043a\u043e\u043b\u043e 9,9 \u0442\u043e\u043a\u0435\u043d\u0430 \u0432 \u0441\u0435\u043a\u0443\u043d\u0434\u0443, \u043d\u0435 \u043e\u0431\u0440\u0430\u0449\u0430\u044f\u0441\u044c \u043a \u0441\u0435\u0440\u0432\u0435\u0440\u0443 \u0438\u043b\u0438 \u0434\u0440\u0443\u0433\u0438\u043c \u0432\u043d\u0435\u0448\u043d\u0438\u043c \u0432\u044b\u0447\u0438\u0441\u043b\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c \u0440\u0435\u0441\u0443\u0440\u0441\u0430\u043c. \u0421\u0433\u0435\u043d\u0435\u0440\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0435 \u043a\u043e\u0440\u043e\u0442\u043a\u0438\u0435 \u0440\u0430\u0441\u0441\u043a\u0430\u0437\u044b \u0432\u044b\u0432\u043e\u0434\u044f\u0442\u0441\u044f \u043d\u0430 \u043f\u043e\u0434\u043a\u043b\u044e\u0447\u0451\u043d\u043d\u044b\u0439 OLED-\u0434\u0438\u0441\u043f\u043b\u0435\u0439.   \u041f\u0440\u043e\u0435\u043a\u0442 esp32-ai \u0438\u0441\u043f\u044b\u0442\u0430\u043d \u043d\u0430 \u043c\u043e\u0434\u0443\u043b\u0435 ESP32-S3 N16R8, \u0440\u0430\u0441\u043f\u043e\u043b\u0430\u0433\u0430\u044e\u0449\u0435\u043c […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":9,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-182824","post","type-post","status-publish","format-standard","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47A 28,9 milli\u00f3 param\u00e9teres nyelvi modell m\u0171k\u00f6dik ESP32-S3-on | ProHoster","description":".","canonical_url":"https:\/\/prohoster.info\/hu\/blog\/news\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"hu_HU","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u042f\u0437\u044b\u043a\u043e\u0432\u0430\u044f \u043c\u043e\u0434\u0435\u043b\u044c \u0441 28,9 \u043c\u043b\u043d \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u043e\u0432 \u0437\u0430\u0440\u0430\u0431\u043e\u0442\u0430\u043b\u0430 \u043d\u0430 ESP32-S3 | ProHoster","og:description":".","og:url":"https:\/\/prohoster.info\/hu\/blog\/news\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2026-08-11T02:54:15+00:00","article:modified_time":"2026-08-12T16:21:41+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"182824","title":null,"description":null,"keywords":null,"keyphrases":{"focus":[],"additional":[]},"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":"default","schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":null,"breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2026-08-11 06:43:49","updated":"2026-08-11 06:43:49","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts\/182824","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/comments?post=182824"}],"version-history":[{"count":1,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts\/182824\/revisions"}],"predecessor-version":[{"id":182897,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts\/182824\/revisions\/182897"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/media?parent=182824"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/categories?post=182824"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/tags?post=182824"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}