--- title: I replaced Grammarly Pro with a local model on my GPU,... description: A free open-source app called Writing Tools works with a local LLM running on the author's own GPU. It sits in the system tray and sends selected text to a... canonical: https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: I replaced Grammarly Pro with a local model on my GPU, and my drafts never leave my PC | daily.dev og:description: A free open-source app called Writing Tools works with a local LLM running on the author's own GPU. It sits in the system tray and sends selected text to a... og:url: https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew og:image: https://api.daily.dev/og/posts/lunmb2jEW.png og:image:alt: I replaced Grammarly Pro with a local model on my GPU, and my drafts never leave my PC 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 replaced Grammarly Pro with a local model on my GPU, and my drafts never leave my PC **[XDA Developers](https://daily.dev/sources/xda-developers)** · 4 min read · 20 upvotes · 1 comments ## Summary A free open-source app called Writing Tools works with a local LLM running on the author's own GPU. It sits in the system tray and sends selected text to a local model via llama-server on localhost, so nothing reaches a third-party server. Pressing Ctrl+Space lets the user proofread text, change tone, or shorten drafts, with Ctrl+Z to undo. In a test with a deliberately messy email containing 24 planted errors, the tool fixed all of them, with only one unrequested sentence reword. Writing Tools also supports Ollama and LM Studio as backends instead of llama.cpp. ## 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 use a local LLM to replace Grammarly for proofreading on Windows? An open-source system tray app called Writing Tools lets you select any text, press Ctrl+Space, and send it to a local model running via llama-server on localhost for proofreading, tone changes, or shortening. The fixed text replaces your selection directly, and Ctrl+Z undoes it. It also supports Ollama and LM Studio as backends instead of llama.cpp. _Developers weighing self-hosted AI tools over subscriptions can track more local-LLM setups like this on daily.dev._ ### Does Grammarly send my text to its servers even if I opt out of AI training? Yes, Grammarly still needs an internet connection to analyze text even after opting out of Product Improvement and Training in account settings, meaning drafts still reach its servers before being processed. This applies to everything from personal emails to corporate communication typed through the assistant. _Privacy-focused developers comparing cloud versus local writing tools can follow more deep dives like this on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@harshilkohli** · 0 upvotes > Hey everyone! > > I've been working on something for quite a while, and I wanted to share it with this community and get your thoughts. > > It's called Oronode Lumina. > > The idea came from a problem I kept running into: AI tools are everywhere, but the experience is fragmented. You use one application for chatting, another for coding, another for running local models, another for automation, and separate tools for image generation, file management, and everything else. > > I wanted to see if I could bring all of that together into one environment. > > That's what I'm building with Lumina: an all-in-one,... ## Similar posts on daily.dev - [I replaced Grammarly with a local LLM, and none of my writing leaves my laptop anymore](https://daily.dev/posts/i-replaced-grammarly-with-a-local-llm-and-none-of-my-writing-leaves-my-laptop-anymore-uaclfv2qb) · XDA Developers · 0 upvotes · 0 comments - [My local LLM is small enough to run on my laptop, but smart enough to replace these 3 subscriptions](https://daily.dev/posts/my-local-llm-is-small-enough-to-run-on-my-laptop-but-smart-enough-to-replace-these-3-subscriptions-7vhtu9law) · XDA Developers · 0 upvotes · 0 comments - [I Tried to Build My Own Grammarly — for the Terminal](https://daily.dev/posts/i-tried-to-build-my-own-grammarly-for-the-terminal-mm5wu6rt1) · Medium · 1 upvotes · 0 comments - [I ditched Google Translate for a local model running on my GPU, and it works offline too](https://daily.dev/posts/i-ditched-google-translate-for-a-local-model-running-on-my-gpu-and-it-works-offline-too-rzild2cil) · XDA Developers · 0 upvotes · 0 comments --- Tags: [#privacy](https://daily.dev/tags/privacy), [#self-hosting](https://daily.dev/tags/self-hosting), [#local-ai](https://daily.dev/tags/local-ai), [#llama-cpp](https://daily.dev/tags/llama-cpp) [View this post on daily.dev](https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew) ```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 replaced Grammarly Pro with a local model on my GPU, and my drafts never leave my PC","url":"https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew"},"datePublished":"2026-11-06T21:04:13.111Z","dateModified":"2026-11-06T21:09:11.170Z","description":"A free open-source app called Writing Tools works with a local LLM running on the author's own GPU. It sits in the system tray and sends selected text to a...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2bec69aa32a1a8c08d8b948c0e6d65e4?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2bec69aa32a1a8c08d8b948c0e6d65e4?_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":1,"discussionUrl":"https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":20},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"privacy,self-hosting,local-ai,llama-cpp","timeRequired":"PT4M"} {"@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 replaced Grammarly Pro with a local model on my GPU, and my drafts never leave my PC"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew","comment":[{"@type":"Comment","text":"Hey everyone!\nI’ve been working on something for quite a while, and I wanted to share it with this community and get your thoughts.\nIt’s called Oronode Lumina.\nThe idea came from a problem I kept running into: AI tools are everywhere, but the experience is fragmented. You use one application for chatting, another for coding, another for running local models, another for automation, and separate tools for image generation, file management, and everything else.\nI wanted to see if I could bring all of that together into one environment.\nThat’s what I’m building with Lumina: an all-in-one, local-first AI desktop workspace.\nHere’s what I’m bringing together:\n\nLocal AI inference: Run compatible GGUF models directly on your own hardware.\nMulti-provider support: Connect different AI providers and local inference engines.\nAI coding workspace: Coding assistance, project workflows, and integrated development tools.\nTerminal and file tools: Work with files, projects, and development environments.\nImage and video generation: Creative generation capabilities within the same workspace.\n3D visual creation: Tools for creating 3D visuals.\nAudio and speech intelligence: Speech processing, OCR, audio understanding, and video understanding.\nGoal Mode: Give Lumina a goal to work toward in the background while you focus on other things.\nAutomation: Build workflows that combine AI, tools, and different actions.\nPlugins and MCP: Extend Lumina with integrations, custom plugins, and MCP servers.\nComputer control: Let AI interact with supported desktop applications and workflows, with permission controls.\nRemote access: Access supported desktop capabilities through a browser.\nAnalytics and personalisation: Customise the interface, models, tools, and overall experience.\n\nWhat about privacy and security?\nLocal-first computing is a major part of the vision. Lumina supports local inference, encrypted local data storage, secure credential handling, and permission-based access to tools and integrations.\nI want users to have meaningful control over their models, data, connections, and workflows. I’m also putting considerable effort into the security architecture, rather than treating security as an afterthought.\nWhy am I building this?\nI don’t think people should need to constantly jump between a dozen applications just to get their AI-powered work done.\nI want Lumina to become a place where AI conversations, development, automation, creative tools, and computing capabilities can work together.\nI’m aiming to release the first version within this month.\nWebsite: https://oronode.dev\nI’d genuinely love to hear what you think.\nIf you work with local AI, coding tools, automation, or desktop AI applications, what would you want to see in a platform like this? What features would make it genuinely useful in your everyday workflow?\nI’m building this independently, and feedback from people who might actually use the product would mean a lot.\nOronode Lumina — One workspace. Your AI. Your control.","datePublished":"2026-11-08T11:40:13.643Z","url":"https://daily.dev/posts/lunmb2jEW#c-ObB9x0KrZ","author":{"@type":"Person","name":"Harshil Kohli","url":"https://daily.dev/harshilkohli","image":"https://lh3.googleusercontent.com/a/ACg8ocJbEhNjfi1UkAr2GzXyDi-uKaBnrmoeQI8F9OlwSHCkhedMQw=s96-c"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-replaced-grammarly-pro-with-a-local-model-on-my-gpu-and-my-drafts-never-leave-my-pc-lunmb2jew#faq","mainEntity":[{"@type":"Question","name":"How can I use a local LLM to replace Grammarly for proofreading on Windows?","acceptedAnswer":{"@type":"Answer","text":"An open-source system tray app called Writing Tools lets you select any text, press Ctrl+Space, and send it to a local model running via llama-server on localhost for proofreading, tone changes, or shortening. The fixed text replaces your selection directly, and Ctrl+Z undoes it. It also supports Ollama and LM Studio as backends instead of llama.cpp. Developers weighing self-hosted AI tools over subscriptions can track more local-LLM setups like this on daily.dev."}},{"@type":"Question","name":"Does Grammarly send my text to its servers even if I opt out of AI training?","acceptedAnswer":{"@type":"Answer","text":"Yes, Grammarly still needs an internet connection to analyze text even after opting out of Product Improvement and Training in account settings, meaning drafts still reach its servers before being processed. This applies to everything from personal emails to corporate communication typed through the assistant. Privacy-focused developers comparing cloud versus local writing tools can follow more deep dives like this on daily.dev."}}]} ```