/ai-ollama-js

Primary LanguageTypeScript

NeAi

This project should help to develop and test much faster with GenAI in ming.

Main functionalities.

RAG.

Get Started

Ollama

Ollama is a tool designed to simplify the installation and management of large language models on local systems.

Install ollama first: https://ollama.com/download/linux

For Linux/Ubuntu/WSL2

curl -fsSL https://ollama.com/install.sh | sh

check that you have ollama installed

ollama --version

# To see possible commands run
ollama --help

We can use many models, and you can use any of them. Here is a list of models.

But we will use llama3.1

So just pull the model over

ollama pull llama3.1

We can check all locally available models by running ollama list.

NAME            ID              SIZE    MODIFIED           
llama3.1:latest 62757c860e01    4.7 GB  About a minute ago      
llama3:latest   a6990ed6be41    4.7 GB  2 months ago

To run ollama use run command

ollama run llama3.1

Now you can communicate with the model 😄 Type /help to see build in commands.

You can leave the terminal chat by using /bye command.