[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:magnitude":3},"\u003Cp align=\"center\">\n  \u003Cpicture>\n    \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"assets\u002Fbrand\u002Ficon-dark.svg\">\u003C\u002Fsource>\n    \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"assets\u002Fbrand\u002Ficon-light.svg\">\u003C\u002Fsource>\n    \u003Cimg alt=\"Magnitude icon\" src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fmagnitudedev\u002Fmagnitude\u002FHEAD\u002Fassets\u002Fbrand\u002Ficon-light.svg\" width=\"120\" \u002F>\n  \u003C\u002Fpicture>\n\u003C\u002Fp>\u003Ch1>Magnitude\u003C\u002Fh1>\u003Cp align=\"center\">\u003Cstrong>Run the best open models for your machine\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fmagnitude.dev\u002Fdownload\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Download-gray?style=flat-square&labelColor=0369a1&logo=data%3Aimage%2Fsvg%2Bxml%3Bbase64%2CPHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHZpZXdCb3g9IjAgMCAyNCAyNCIgZmlsbD0ibm9uZSIgc3Ryb2tlPSIjZmZmZmZmIiBzdHJva2Utd2lkdGg9IjIuMjUiIHN0cm9rZS1saW5lY2FwPSJyb3VuZCIgc3Ryb2tlLWxpbmVqb2luPSJyb3VuZCI%2BPHBhdGggZD0iTTEyIDN2MTIiLz48cGF0aCBkPSJtNyAxMCA1IDUgNS01Ii8%2BPHBhdGggZD0iTTQgMTd2MmEyIDIgMCAwIDAgMiAyaDEyYTIgMiAwIDAgMCAyLTJ2LTIiLz48L3N2Zz4%3D\" alt=\"Download Magnitude\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdocs.magnitude.dev\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%93%95-Docs-0369a1?style=flat-square&labelColor=0369a1&color=gray\" alt=\"Documentation\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002FEHt48pPWdC\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Discord-gray?style=flat-square&logo=discord&logoColor=white&labelColor=5865F2\" alt=\"Discord\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fx.com\u002Fusemagnitude\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Twitter-gray?style=flat-square&logo=x&logoColor=white&labelColor=000000\" alt=\"Follow Magnitude on Twitter\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fmagnitudedev\u002Fmagnitude\u002Fstargazers\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagnitudedev\u002Fmagnitude\" alt=\"GitHub Repo stars\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Cp>Magnitude is the open source inference engine for the hardware you already own. It profiles your machine, recommends the best open models for it, and tunes them for your exact hardware. One click connects your favorite agent (Pi, OpenCode, Hermes, etc.). Works on Apple Silicon, NVIDIA, AMD, or nothing but a CPU.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>\u003Ca href=\"https:\u002F\u002Fmagnitude.dev\u002Fdownload\" rel=\"nofollow ugc noopener\">Download Magnitude for macOS, Windows, or Linux\u003C\u002Fa>\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cp>⭐ Help us reach more developers and grow the Magnitude community. Star this repo!\u003C\u002Fp>\n\u003Cp>\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F8317d05b-8a6e-40e0-b45d-81011ecbc329\" rel=\"nofollow ugc noopener\">https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F8317d05b-8a6e-40e0-b45d-81011ecbc329\u003C\u002Fa>\u003C\u002Fp>\n\u003Ch2>Get started\u003C\u002Fh2>\n\u003Col>\n\u003Cli>\u003Ca href=\"https:\u002F\u002Fmagnitude.dev\u002Fdownload\" rel=\"nofollow ugc noopener\">Download Magnitude\u003C\u002Fa>, install it, and open the app.\u003C\u002Fli>\n\u003Cli>Choose a recommended model in \u003Cstrong>Discover\u003C\u002Fstrong> and download it.\u003C\u002Fli>\n\u003Cli>Connect your agent in \u003Cstrong>Connections\u003C\u002Fstrong> and start using it.\u003C\u002Fli>\n\u003C\u002Fol>\n\u003Cp>The desktop app includes the \u003Ccode>magnitude\u003C\u002Fcode> CLI. No separate installation is needed.\u003C\u002Fp>\n\u003Ch2>Why Magnitude?\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Cstrong>Knows your machine:\u003C\u002Fstrong> profiles your hardware and estimates tok\u002Fs before you download\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Recommends the best models:\u003C\u002Fstrong> ranked by speed, accuracy, intelligence, and memory\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Tuned end to end:\u003C\u002Fstrong> speculative decoding and more, all set for your hardware\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Works with your agent:\u003C\u002Fstrong> one click to connect Pi, OpenCode, Hermes, and more\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Free to run:\u003C\u002Fstrong> no token costs, API keys, or rate limits\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Fully private and offline:\u003C\u002Fstrong> models, prompts, and files stay on your machine\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Models on demand:\u003C\u002Fstrong> loaded on request, unloaded when idle or memory fills\u003C\u002Fli>\n\u003Cli>\u003Cstrong>Open source:\u003C\u002Fstrong> Apache 2.0, yours to modify\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>FAQ\u003C\u002Fh2>\n\u003Ch3>What is Magnitude?\u003C\u002Fh3>\n\u003Cp>An open source inference engine optimized for consumer hardware. The desktop app profiles your machine, recommends the best models for it, then downloads, tunes, and runs them. One click connects the agent you already use.\u003C\u002Fp>\n\u003Ch3>How does it know what my machine can run?\u003C\u002Fh3>\n\u003Cp>Magnitude profiles your hardware and estimates tok\u002Fs for every model in the catalog before you download anything. It ranks them by speed, accuracy, intelligence, and memory so you can pick.\u003C\u002Fp>\n\u003Ch3>How is this different from Ollama or LM Studio?\u003C\u002Fh3>\n\u003Cp>They run whatever model you pick. Magnitude helps you pick. It estimates how every model and quant will perform on your machine before you download, then tunes the one you choose for your exact hardware, from context size to speculative de\u003C\u002Fp>\n",1790633830019]