Local Ollama Docker setup

🎯 Overview

This project contains instructions on how to locally run Ollama together with Ollama WebUI.

📦 Requirements

In order to be able to create Weaviate one needs at least.

1. Docker

💡 Running using Docker compose

In order to run the setup locally running only docker compose use this comman to run the environment:

docker compose up -d

In order to pull ollama models issue:

docker exec -i ollama ollama pull <model_name>

for example if you want to download a latest multilingual large language model from Cohere - Aya issue:

docker exec -i ollama ollama pull aya

more models can be found using Ollama library site.

Ollama WebUI will start woking on port 8080, just type localhost:8080 in your browser. Please note that at the very first login attempt you would need to create an account with which you can then login to Ollama WebUI.