LLM-automator

Huggingface Backend

Installation

Get HF_TOKEN from https://huggingface.co/settings/tokens

Edit llm-automator.workflow with the Automator app, replace HF_TOKEN with your token.

You may also edit MODEL="mistralai/Mixtral-8x7B-Instruct-v0.1" to use a different LLM from huggingface.co/models

Install the workflow. This essentially copies the workflow to ~/Library/Services/llm-automator.workflow.

Go to System Preferences -> Keyboard -> Shortcuts -> Services -> General -> llm-automator, and set the shortcut to whatever you want, in my case I set it to ⌃⌥⌘M.

Usage

In any OSX application, select some text, and press the shortcut you set in step 3. The selected text will be replaced with the generated text from the model.

Ollama Backend

Install ollama

Run in the background:

ollama run mistral

Install llm-automator-ollama.workflow the same way as above, no need to edit and add token, however you may edit MODEL="mistral", in case you want to use a different model


asdasdsaas.mp4