--- title: I stopped uploading my notes to ChatGPT and NotebookLM... description: A developer shares their workflow for replacing cloud-based AI note-taking tools (ChatGPT, NotebookLM) with a fully local setup. The stack uses embedding... canonical: https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: I stopped uploading my notes to ChatGPT and NotebookLM after setting up my local LLMs | daily.dev og:description: A developer shares their workflow for replacing cloud-based AI note-taking tools (ChatGPT, NotebookLM) with a fully local setup. The stack uses embedding... og:url: https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h og:image: https://api.daily.dev/og/posts/xrWcWWA8h.png og:image:alt: I stopped uploading my notes to ChatGPT and NotebookLM after setting up my local LLMs 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. # I stopped uploading my notes to ChatGPT and NotebookLM after setting up my local LLMs **[XDA Developers](https://daily.dev/sources/xda-developers)** · 5 min read · 1 upvotes · 0 comments ## Summary A developer shares their workflow for replacing cloud-based AI note-taking tools (ChatGPT, NotebookLM) with a fully local setup. The stack uses embedding models (Nomic Embed Text, Qwen3-Embedding) for RAG pipelines, mixture-of-experts LLMs (Gemma-4-26B, Qwen3-35B) for inference, Blinko for quick notes and personal knowledge management, and Open Notebook as a self-hosted NotebookLM alternative for research. The author highlights that low-parameter embedding models run on modest hardware like a Raspberry Pi, and MoE models fit on VRAM-constrained GPUs, making the entire pipeline viable without high-end hardware. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What embedding model can run on a Raspberry Pi for local RAG note-taking? Nomic Embed Text, specifically the V1.5 variant, is light enough to run on a Raspberry Pi while still working well in retrieval-augmented generation pipelines for note-taking apps. For higher precision needs, Qwen3-Embedding (8B) is a heavier alternative, though low-parameter embedding models are generally sufficient for typical note-taking workflows. _daily.dev surfaces practical setups like this for developers building local RAG pipelines on constrained hardware._ ### Why use mixture-of-experts LLMs instead of dense models for local inference on a consumer GPU? Mixture-of-experts models run the attention and router mechanisms on VRAM while offloading rarely-used expert layers to CPU and system memory, letting them fit on VRAM-constrained GPUs without major performance loss. Examples include Gemma-4-26B-A4B running at 15 tokens per second on a GTX card, and GPT-OSS-20B as a faster alternative. _Developers picking local models for limited hardware can track real-world benchmarks like this on daily.dev._ ### What is a self-hosted alternative to NotebookLM for research notes? Open Notebook functions as a self-hosted variant of NotebookLM, storing PDFs, blog posts, and custom notes for academic and research use while supporting RAG search across many sources. It can also convert documentation into summarized podcasts using locally hosted text-to-speech and speech-to-text models via Speaches. _Those weighing self-hosted versus cloud research tools can follow comparisons like this on daily.dev._ ## Similar posts on daily.dev - [I stopped uploading research to NotebookLM, and my local LLM does everything Google's version promised](https://daily.dev/posts/i-stopped-uploading-research-to-notebooklm-and-my-local-llm-does-everything-google-s-version-promis-nagbmrwtp) · XDA Developers · 0 upvotes · 0 comments - [I replaced NotebookLM with this free tool that uses my local LLMs](https://daily.dev/posts/i-replaced-notebooklm-with-this-free-tool-that-uses-my-local-llms-tao99dqvm) · XDA Developers · 1 upvotes · 0 comments - [Running local models is good now](https://daily.dev/posts/running-local-models-is-good-now-q20vqhe78) · Lobsters · 1 upvotes · 0 comments - [I switched from NotebookLM to this open-source tool, and it does what Google won't](https://daily.dev/posts/i-switched-from-notebooklm-to-this-open-source-tool-and-it-does-what-google-won-t-ojzoxs5xr) · XDA Developers · 0 upvotes · 0 comments --- Tags: [#rag](https://daily.dev/tags/rag), [#self-hosting](https://daily.dev/tags/self-hosting), [#local-ai](https://daily.dev/tags/local-ai), [#embeddings](https://daily.dev/tags/embeddings) [View this post on daily.dev](https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h) ```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":"I stopped uploading my notes to ChatGPT and NotebookLM after setting up my local LLMs","url":"https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h"},"datePublished":"2026-07-20T19:31:48.206Z","dateModified":"2026-09-14T08:20:38.623Z","description":"A developer shares their workflow for replacing cloud-based AI note-taking tools (ChatGPT, NotebookLM) with a fully local setup. The stack uses embedding...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f83e93af6d6078cf44f2de028db64267?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/f83e93af6d6078cf44f2de028db64267?_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/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"rag,self-hosting,local-ai,embeddings","timeRequired":"PT5M"} {"@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":"I stopped uploading my notes to ChatGPT and NotebookLM after setting up my local LLMs"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-stopped-uploading-my-notes-to-chatgpt-and-notebooklm-after-setting-up-my-local-llms-xrwcwwa8h#faq","mainEntity":[{"@type":"Question","name":"What embedding model can run on a Raspberry Pi for local RAG note-taking?","acceptedAnswer":{"@type":"Answer","text":"Nomic Embed Text, specifically the V1.5 variant, is light enough to run on a Raspberry Pi while still working well in retrieval-augmented generation pipelines for note-taking apps. For higher precision needs, Qwen3-Embedding (8B) is a heavier alternative, though low-parameter embedding models are generally sufficient for typical note-taking workflows. daily.dev surfaces practical setups like this for developers building local RAG pipelines on constrained hardware."}},{"@type":"Question","name":"Why use mixture-of-experts LLMs instead of dense models for local inference on a consumer GPU?","acceptedAnswer":{"@type":"Answer","text":"Mixture-of-experts models run the attention and router mechanisms on VRAM while offloading rarely-used expert layers to CPU and system memory, letting them fit on VRAM-constrained GPUs without major performance loss. Examples include Gemma-4-26B-A4B running at 15 tokens per second on a GTX card, and GPT-OSS-20B as a faster alternative. Developers picking local models for limited hardware can track real-world benchmarks like this on daily.dev."}},{"@type":"Question","name":"What is a self-hosted alternative to NotebookLM for research notes?","acceptedAnswer":{"@type":"Answer","text":"Open Notebook functions as a self-hosted variant of NotebookLM, storing PDFs, blog posts, and custom notes for academic and research use while supporting RAG search across many sources. It can also convert documentation into summarized podcasts using locally hosted text-to-speech and speech-to-text models via Speaches. Those weighing self-hosted versus cloud research tools can follow comparisons like this on daily.dev."}}]} ```