AI LLM

approved

by Sparky4567

Integrate local machine learning (OLLAMA) functionality into your notes, enhancing their capabilities.

★ 63 stars↓ 7,388 downloadsUpdated 1y agoApache-2.0
View on GitHub

Obsidian AI plugin

This plugin was developed to make a better way to use Local LLM models with Obsidian.

In order to use the plugin

  • Download/Clone the plugin into your plugins folder

cd ./obsidian/plugins
git clone https://github.com/Sparky4567/obsidian_ai_plugin.git
cd obsidian_ai_plugin
npm install
npm run build
open Obsidian app
enable community plugin support
enable LLM plugin
choose a module within the settings tab
  • Ensure that you have Ollama installed

https://ollama.com/download

Read documentation accordingly.

  • Ensure that Ollama model is running in the background

ollama run tinyllama

(Example)
  • If you downloaded this plugin from GitHub repo, copy it to your .obsidian/plugins, don't forget to run npm install within the plugins directory
npm install

to install all needed dependencies.

  • Ensure that the plugin is activated
  • Choose the right endpoint and model in plugins settings
  • Write something into editor field (Simple text)
  • Select the text with your mouse
  • Press CTRL+P after selection
  • Type in ASK LLM and choose your wanted command (There aren't many at the moment)
  • Press Enter to confirm
  • Wait for a while to get the result

Your text will be changed with the text from LLM (Default is tinyllama)

If you have any questions related to the plugin or want to extend the functionality, write an email to admin@artefaktas.eu and I will try to respond as soon, as I can.

Recommendations

  • A laptop with at least 8GB of RAM and a decent processor (for local usage)

Want to support the project ?

ko-fi

wakatime

For plugin developers

Search results and similarity scores are powered by semantic analysis of your plugin's README. If your plugin isn't appearing for searches you'd expect, try updating your README to clearly describe your plugin's purpose, features, and use cases.