--- title: My phone runs an AI assistant entirely offline, and I... description: Local LLM apps on mobile rarely support document uploads, forcing users to choose between convenience and privacy when handling sensitive files like bank... canonical: https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: My phone runs an AI assistant entirely offline, and I stopped uploading my documents to cloud chatbots | daily.dev og:description: Local LLM apps on mobile rarely support document uploads, forcing users to choose between convenience and privacy when handling sensitive files like bank... og:url: https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze og:image: https://api.daily.dev/og/posts/jXd4ME3Ze.png og:image:alt: My phone runs an AI assistant entirely offline, and I stopped uploading my documents to cloud chatbots 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. # My phone runs an AI assistant entirely offline, and I stopped uploading my documents to cloud chatbots **[XDA Developers](https://daily.dev/sources/xda-developers)** · 5 min read · 0 upvotes · 0 comments ## Summary Local LLM apps on mobile rarely support document uploads, forcing users to choose between convenience and privacy when handling sensitive files like bank statements or medical results. Noema, a free open-source iOS app, fills this gap by running a full retrieval pipeline on-device: it downloads an embedding model (Qwen3 Embedding 0.6B), chunks and embeds documents locally, and supports PDF, EPUB, text, and markdown import via the Files app. It runs models in GGUF, MLX, Liquid AI SLM, ExecuTorch, and CoreML formats, plus Apple's Foundation Model on Apple Intelligence devices. A built-in RAM advisor estimates whether a model fits before loading. Trade-offs include slower inference (a few tokens per second on larger models) and weaker reasoning compared to frontier cloud models, but the author considers full local document privacy worth the performance hit. Android users are pointed to LocalRAG as an equivalent. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I get an offline AI app on my phone to read and answer questions about my PDF documents without uploading them to the cloud? Noema, a free open-source iOS app, supports document upload with a full on-device RAG pipeline: it chunks and embeds PDFs, EPUBs, plain text, and markdown files locally using an embedding model (Qwen3 Embedding 0.6B, roughly 640MB), then lets you query them with no cloud calls. Android users can use LocalRAG as an equivalent. _daily.dev surfaces practical local-AI setups for developers who want to build or evaluate on-device RAG workflows._ ### What model formats does the Noema local AI app support on iOS? Noema supports five model formats: GGUF, MLX, Liquid AI's SLM, ExecuTorch, and CoreML, and it also works with Apple's Foundation Model on Apple Intelligence-enabled devices. It includes a RAM advisor that estimates whether a model will fit on the device before loading, avoiding failed downloads and crashes. _Developers comparing on-device model runtimes can track format and tooling shifts like this on daily.dev._ ### Why don't most local LLM apps like PocketPal or Google AI Edge Gallery support document uploads? Mobile local LLM apps such as PocketPal and Google's AI Edge Gallery focus on running chat models on-device but lack native document ingestion, requiring users to manually copy-paste text instead of attaching files. Neither includes a built-in retrieval-augmented generation system, which is why apps like Noema and LocalRAG that add on-device embedding and chunking stand out. _daily.dev helps developers evaluating local AI tooling stay on top of which apps add real document workflows._ ## Similar posts on daily.dev - [I ran a full LLM on my phone with no internet, and it's more useful than I expected](https://daily.dev/posts/i-ran-a-full-llm-on-my-phone-with-no-internet-and-it-s-more-useful-than-i-expected-nndli0pwm) · XDA Developers · 0 upvotes · 0 comments - [I paired a local LLM with Obsidian on my phone, and it's the productivity boost I didn't expect](https://daily.dev/posts/i-paired-a-local-llm-with-obsidian-on-my-phone-and-it-s-the-productivity-boost-i-didn-t-expect-kpxgxdofh) · XDA Developers · 0 upvotes · 0 comments - [Medium](https://daily.dev/posts/medium-kz1oa1nub) · Medium · 1 upvotes · 0 comments --- Tags: [#privacy](https://daily.dev/tags/privacy), [#rag](https://daily.dev/tags/rag), [#local-ai](https://daily.dev/tags/local-ai) [View this post on daily.dev](https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze) ```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":"My phone runs an AI assistant entirely offline, and I stopped uploading my documents to cloud chatbots","url":"https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze"},"datePublished":"2026-08-24T20:01:48.539Z","dateModified":"2026-08-24T20:05:01.437Z","description":"Local LLM apps on mobile rarely support document uploads, forcing users to choose between convenience and privacy when handling sensitive files like bank...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e21f9619d009e592b90ec1df27b9edb2?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e21f9619d009e592b90ec1df27b9edb2?_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/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"privacy,rag,local-ai","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":"My phone runs an AI assistant entirely offline, and I stopped uploading my documents to cloud chatbots"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/my-phone-runs-an-ai-assistant-entirely-offline-and-i-stopped-uploading-my-documents-to-cloud-chatbo-jxd4me3ze#faq","mainEntity":[{"@type":"Question","name":"How can I get an offline AI app on my phone to read and answer questions about my PDF documents without uploading them to the cloud?","acceptedAnswer":{"@type":"Answer","text":"Noema, a free open-source iOS app, supports document upload with a full on-device RAG pipeline: it chunks and embeds PDFs, EPUBs, plain text, and markdown files locally using an embedding model (Qwen3 Embedding 0.6B, roughly 640MB), then lets you query them with no cloud calls. Android users can use LocalRAG as an equivalent. daily.dev surfaces practical local-AI setups for developers who want to build or evaluate on-device RAG workflows."}},{"@type":"Question","name":"What model formats does the Noema local AI app support on iOS?","acceptedAnswer":{"@type":"Answer","text":"Noema supports five model formats: GGUF, MLX, Liquid AI's SLM, ExecuTorch, and CoreML, and it also works with Apple's Foundation Model on Apple Intelligence-enabled devices. It includes a RAM advisor that estimates whether a model will fit on the device before loading, avoiding failed downloads and crashes. Developers comparing on-device model runtimes can track format and tooling shifts like this on daily.dev."}},{"@type":"Question","name":"Why don't most local LLM apps like PocketPal or Google AI Edge Gallery support document uploads?","acceptedAnswer":{"@type":"Answer","text":"Mobile local LLM apps such as PocketPal and Google's AI Edge Gallery focus on running chat models on-device but lack native document ingestion, requiring users to manually copy-paste text instead of attaching files. Neither includes a built-in retrieval-augmented generation system, which is why apps like Noema and LocalRAG that add on-device embedding and chunking stand out. daily.dev helps developers evaluating local AI tooling stay on top of which apps add real document workflows."}}]} ```