DEV Community

Cover image for Devika - Install in local
Kiran J
Kiran J

Posted on

Devika - Install in local

I have installed Devika locally using docker.
As a prerequisit, I have already installed docker in my ubuntu.

There are three main components in Devika,

  1. Frontend server which runs on port 3000
  2. Backend Server
  3. Ollama and ollama LLM models which runs on port 11434

Devika can be use different non-open source LLM Models like OpenAI (chatgpt), ClaudeAI etc..

Here are step by step instructions.
Clone repository in your local directory using terminal.

git clone https://github.com/stitionai/devika.git
cd devika
Enter fullscreen mode Exit fullscreen mode

copy config.toml and edit to update API keys.

cp sample.config.toml config.toml
sudo nano config.toml
Enter fullscreen mode Exit fullscreen mode

Now, you can start installing process using docker compose. Devika's repository already contains docker compose file and no need to change anything.

sudo docker compose up

Now in the terminal, you can see the devika's containers starting.

Image description

finally you can see the port devika is running

Image description

You can go to browser and run localhost:3000.

Top comments (0)