{"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\/da\/blog\/news\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","title":{"rendered":"Sproglige modellen med 28,9 millioner parametre k\u00f8rte p\u00e5 ESP32-S3","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

<\/i><\/a>
\nUdvikleren Vyacheslav Serbov, der bruger pseudonymet slvDev, demonstreret<\/a> har udviklet en helt lokal tekstgenerator p\u00e5 mikrokontrolleren ESP32-S3. Den sprogmodel, han har skabt, indeholder 28,9 millioner parametre og genererer cirka 9,9 tokens pr. sekund uden at g\u00f8re brug af en server eller andre eksterne computerressourcer. De genererede korte fort\u00e6llinger vises p\u00e5 en tilsluttet OLED-sk\u00e6rm.<\/p>\n

 <\/p>\n

Projekt esp32-ai<\/a> er testet p\u00e5 ESP32-S3 N16R8-modulet, som har 512 KB indbygget SRAM, 8 MB PSRAM og 16 MB flashhukommelse. Quantiseret til fire bit fylder modellen 14,9 MB. Kildekoden til runtime-milj\u00f8et i C, tr\u00e6nings- og kvantiseringsv\u00e6rkt\u00f8jerne, firmware, tests og dens upload-scripts er offentliggjort i depotet.<\/p>\n

 <\/p>\n

De angivne 28,9 millioner parametre b\u00f8r ikke sammenlignes direkte med antallet af parametre for traditionelle desktop LLM'er. If\u00f8lge den detaljerede rapport fra udvikleren,<\/a>best\u00e5r modellen af et t\u00e6t beregningskerne med cirka 559 tusinde parametre, et udgangslag med 3,1 millioner parametre og et lagdelt vektortabel med cirka 25 millioner parametre. Det er is\u00e6r det sidste, der giver den prim\u00e6re del af den formelle st\u00f8rrelse af modellen, men behandles ikke helt ved genereringen af hvert token.<\/p>\n

 <\/p>\n

Tre niveauer af hukommelse anvendes til at h\u00e5ndtere modellen. De mest brugte data findes i den interne SRAM, udgangslaget, KV-cache og midlertidige buffere er i PSRAM, mens den 25-millioners tabel forbliver i flashhukommelsen og er tilg\u00e6ngelig gennem kortl\u00e6gning til adressens rum. For hvert token bliver der kun l\u00e6st omkring seks linjer med et samlet volumen p\u00e5 cirka 450 byte fra den.<\/p>\n

 <\/p>\n

S\u00e5dan en enhed er baseret p\u00e5 Per-Layer Embeddings-teknologien, anvendt af Google i Gemma 3n. If\u00f8lge Google-dokumentationen,<\/a>kan PLE-parametrene opbevares uden for modelens arbejdshukommelse og tilf\u00f8jes i outputprocessen, mens de enkelte lag behandles. I esp32-ai er denne tilgang overf\u00f8rt til mikrokontrollerens hukommelseshierarki, hvor hurtig SRAM er s\u00e6rligt begr\u00e6nset.<\/p>\n

 <\/p>\n

Den f\u00f8rste korrekte version af det udf\u00f8rte C-k\u00f8rselmilj\u00f8 gav kun 0,57 tokens per sekund. Efter at have placeret outputlaget i PSRAM, skiftet til 8-bit aktiveringer, fordelt beregninger mellem to Xtensa LX7-kerner og andre optimeringer, blev latenstiden reduceret til 94,9 ms pr. modeltrin. Den endelige m\u00e5lte hastighed n\u00e5ede 9,88 tokens per sekund med hensyn til den samlede genereringsproces. Den prim\u00e6re begr\u00e6nsning er nu PSRAM-b\u00e5ndbredden ved l\u00e6sning af outputlaget.<\/p>\n

 <\/p>\n

Modellen er tr\u00e6net p\u00e5 et syntetisk datas\u00e6t TinyStories<\/a>, best\u00e5ende af enkle engelsksprogede historier med et begr\u00e6nset ordforr\u00e5d. Derfor er den i stand til at forts\u00e6tte s\u00e6tninger og skabe sm\u00e5 sammenh\u00e6ngende historier, men den er ikke designet til at besvare sp\u00f8rgsm\u00e5l, udf\u00f8re instruktioner, programmere eller gengive faktiske viden. Udvikleren betragter arbejdet f\u00f8rst og fremmest som en demonstration af en effektiv modell\u00e6gningsplads i mikrocontrollerens hukommelse, snarere end som en f\u00e6rdig autonom chatbot.<\/p>\n

 <\/p>\n

I den aktuelle dokumentation understreger forfatteren, at esp32-ai er et uafh\u00e6ngigt projekt. Projekter llama2.c<\/a> Andrej Karpaty og den tidligere lancering af den 260.000 parametre model p\u00e5 ESP32-S3 n\u00e6vnes kun som pejlem\u00e6rker og eksempler p\u00e5 tidligere arbejde: koden, kontrolpunkterne og metoden er ikke direkte l\u00e5nt fra dem. Kildekoden til esp32-ai distribueres under MIT-licensen<\/a>.<\/p>\n

 <\/p>\n

Kilde: 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\udd47Sprogmodellen med 28,9 millioner parametre fungerer p\u00e5 ESP32-S3 | ProHoster","description":".","canonical_url":"https:\/\/prohoster.info\/da\/blog\/news\/yazykovaya-model-s-289-mln-parametrov-zarabotala-na-esp32-s3","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"da_DK","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\/da\/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\/da\/wp-json\/wp\/v2\/posts\/182824","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/comments?post=182824"}],"version-history":[{"count":1,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts\/182824\/revisions"}],"predecessor-version":[{"id":182897,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts\/182824\/revisions\/182897"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/media?parent=182824"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/categories?post=182824"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/tags?post=182824"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}