Companion LLM
Companion is your private and personal AI chat that lives on your desktop. Here are the steps to get it setup, and to deploy your first custom model to chat with in companion.
Install Companion and Navigator
To run Companion and deploy flows to it, you'll need to install and set up Navigator first.
- Open Navigator, create your account, and sign in.
Dependencies will install. Accept the x-code pop-up when the set-up gets to Step 6. It may be hidden behind another window.
- Once in Navigator, go to the Clusters Tab and set up a cluster.
- Host a New Cluster
- Add a Name and Description
- Click into your new cluster and add a Node
- Select your device, and any other devices you are running navigator on that you would like to host or use your models and give the device a name.
For an in-depth guide on clusters head to Deploying & Using Clusters
- Host a New Cluster
- Open Companion
- Login with the same credentials as your Navigator Account
- Wait for the included LLM to install. This may take some time.
- Once the Status dot next to the Icon turns green, start chatting.
Threads and Attachments
Attachments are created automatically by Companion.
Click the attachment to open it in the attachment panel.
Flip between attachments using the back arrow on the panel.
New threads can also be created using the left panel.
Create a Custom Chatbot
Once you have a model you are happy with save the flow, and Deploy it to your cluster. Once its Active - the model will show up in the left side bar of companion.
Compatible Elements
For an in-depth guide on fine tuning a model with custom information - Creating a Local Expert LLM
Using your Custom Model
When models are deployed to your cluster they will automatically appear in the left sidebar of companion. Each icon represents a different Deployment in Navigator. You can switch between models and have different threads with each model using the left side bar.