--- title: Someone squeezed a 28.9M LLM onto an ESP32-S3, and so... description: A developer managed to run a 28.9M-parameter LLM on an ESP32-S3 microcontroller at ~9 tokens/second. The key technique leverages Gemma's per-layer embeddings:... canonical: https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you | daily.dev og:description: A developer managed to run a 28.9M-parameter LLM on an ESP32-S3 microcontroller at ~9 tokens/second. The key technique leverages Gemma's per-layer embeddings:... og:url: https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva og:image: https://api.daily.dev/og/posts/A5f9lZPVa.png og:image:alt: Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you **[XDA Developers](https://daily.dev/sources/xda-developers)** · 2 min read · 2 upvotes · 0 comments ## Summary A developer managed to run a 28.9M-parameter LLM on an ESP32-S3 microcontroller at ~9 tokens/second. The key technique leverages Gemma's per-layer embeddings: the 25M-parameter embedding table is stored in flash memory (memory-mapped XIP) and only ~6 rows (~450 bytes) are read per token, while only a ~560K dense core needs fast RAM. The result is a fully on-chip AI capable of generating coherent stories. The project is open source and available on GitHub. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [28.9M-parameter LLM runs locally on ESP32-S3 at 9 tokens/s](https://daily.dev/posts/28-9m-parameter-llm-runs-locally-on-esp32-s3-at-9-tokens-s-xmajabusu) · CNX Software · 0 upvotes · 0 comments - [Someone wired up seven ESP32s to create a ~0.4B LLM, and so can you](https://daily.dev/posts/someone-wired-up-seven-esp32s-to-create-a-0-4b-llm-and-so-can-you-xwbd0nns2) · XDA Developers · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#iot](https://daily.dev/tags/iot), [#embedded](https://daily.dev/tags/embedded), [#edge-ai](https://daily.dev/tags/edge-ai) [View this post on daily.dev](https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva) ```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":"Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you","url":"https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva"},"datePublished":"2026-07-24T01:47:16.187Z","dateModified":"2026-07-24T04:23:02.249Z","description":"A developer managed to run a 28.9M-parameter LLM on an ESP32-S3 microcontroller at ~9 tokens/second. The key technique leverages Gemma's per-layer embeddings:...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/75734d7141e28a767a9658281e67daae?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/75734d7141e28a767a9658281e67daae?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","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":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/someone-squeezed-a-28-9m-llm-onto-an-esp32-s3-and-so-can-you-a5f9lzpva","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,iot,embedded,edge-ai","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"Someone squeezed a 28.9M LLM onto an ESP32-S3, and so can you"}]} ```