Setup Ollama for BasicModel

One of the easiest way to host an open source model and connect to Linguly is to host a model from Ollama library in Coolify.

How to?

1. Download and Run a Model Using Ollama

1. Add ollama-with-open-webui in Coolify resources to a docker network same as your Linguly service.
2. Go to the terminal and connect to ollama api
3. Use the following command to download a desired model: ollama pull llama3.2:3b
   1. To check the result you can use ollama list
   2. If you want to run and test the model in terminal: ollama run llama3.2:3b

For Local Tests

You can install ollama locally from here and then pull a model similarly.