--- title: If you want, you can't just run a 1T model from SSD, but... description: Running large language models locally is already too slow for practical use even on high-end Apple Silicon hardware like Mx chips. Even relatively small models... canonical: https://daily.dev/posts/if-you-want-you-can-t-just-run-a-1t-model-from-ssd-but-a-10t-model-from-manual-pen-and-paper-math--fforv28cp twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: If you want, you can't just run a 1T model from SSD, but a 10T model from manual pen and paper math. The fact is that even small LLMs like Qwen 3.5 35B are already almost too slow for serious usage in high end Mx Apple Silicon. Running big LLMs slow is nice, but useless. | daily.dev og:description: Running large language models locally is already too slow for practical use even on high-end Apple Silicon hardware like Mx chips. Even relatively small models... og:url: https://daily.dev/posts/if-you-want-you-can-t-just-run-a-1t-model-from-ssd-but-a-10t-model-from-manual-pen-and-paper-math--fforv28cp og:image: https://api.daily.dev/og/posts/FfORv28Cp.png og:image:alt: If you want, you can't just run a 1T model from SSD, but a 10T model from manual pen and paper math. The fact is that even small LLMs like Qwen 3.5 35B are already almost too slow for serious usage in high end Mx Apple Silicon. Running big LLMs slow is nice, but useless. og:image:width: 1200 og:image:height: 630 og:locale: en --- [antirez](https://daily.dev/sources/antirez) [Read on](https://api.daily.dev/r/FfORv28Cp) Mar 24 • From x.com  antirez @antirez If you want, you can't just run a 1T model from SSD, but a 10T model from manual pen and paper math. The fact is that even small LLMs like Qwen 3.5 35B are already almost too slow for serious usage in high end Mx Apple Silicon. Running big LLMs slow is nice, but useless. Comment Bookmark Copy Share your thoughts Post Share this post [](https://daily.dev/sources/antirez) [antirez](https://daily.dev/sources/antirez "https://daily.dev/sources/antirez") 7 Followers • 4 Upvotes #### Would you recommend this post? Copy link Slack WhatsApp Facebook X New Squad Copy linkShare with your friends ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"If you want, you can't just run a 1T model from SSD, but a 10T model from manual pen and paper math. The fact is that even small LLMs like Qwen 3.5 35B are already almost too slow for serious usage in high end Mx Apple Silicon. Running big LLMs slow is nice, but useless.","url":"https://daily.dev/posts/if-you-want-you-can-t-just-run-a-1t-model-from-ssd-but-a-10t-model-from-manual-pen-and-paper-math--fforv28cp","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/if-you-want-you-can-t-just-run-a-1t-model-from-ssd-but-a-10t-model-from-manual-pen-and-paper-math--fforv28cp"},"datePublished":"2026-03-24T16:55:26.285Z","dateModified":"2026-03-24T16:55:36.916Z","description":"Running large language models locally is already too slow for practical use even on high-end Apple Silicon hardware like Mx chips. Even relatively small models...","image":"https://media.daily.dev/image/upload/s--HRgLpUt6--/f_auto/v1722860399/public/Placeholder%2003","thumbnailUrl":"https://media.daily.dev/image/upload/s--HRgLpUt6--/f_auto/v1722860399/public/Placeholder%2003","isAccessibleForFree":true,"articleSection":"antirez","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"antirez","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/d8d91fc3b30b407d894efa6d6c02e51f","url":"https://daily.dev/sources/antirez"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/if-you-want-you-can-t-just-run-a-1t-model-from-ssd-but-a-10t-model-from-manual-pen-and-paper-math--fforv28cp","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":""} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"antirez","item":"https://daily.dev/sources/antirez"},{"@type":"ListItem","position":3,"name":"If you want, you can't just run a 1T model from SSD, but a 10T model from manual pen and paper math. The fact is that even small LLMs like Qwen 3.5 35B are already almost too slow for serious usage in high end Mx Apple Silicon. Running big LLMs slow is nice, but useless."}]} ```