hocki1 / offline_chatgpt

Offline ChatGPT using Large Language Model (LLM)

Offline ChatGPT using Large Language Model (LLM)

Here we will see, how to run a ChatGPT-like LLM on a local machine without internet. This I have tested on macOS(13.1)

Prerequisite

• Python 3.9.7
• NodeJS version 18.0 or later
• Pipenv

Build with

• Python 3.9.7
• Alpaca model

What is Alpaca and LLaMA?

Alpaca is a small AI language model developed by a group of computer scientists at Stanford University. Alpaca is fine-tuned on LLaMa, Meta’s large language model which recently leaked online.

Follow the below steps to install and start Offline ChatGPT

1. Here we are using Alpaca 7B LLM model (around 4.2 GB disk space). If we want to install the Alpaca 13B model, then we need to replace 7B with 13B. Note the larger model needs 8.1 GB of space. We can use the below command to install alpaca model.

npx dalai alpaca install 7B

2. Run the below command to start the Web Server

npx dalai serve

3. Open the Web Browser at the below URL

http://localhost:3000

4. We can now start using this Offline ChatGPT-like language model on our PC without internet connectivity. Replace PROMPT with the query we wanna try and click on Go.

5. Optional: We can also install LLaMa models. LLaMA models are much larger. Its 7B model takes up to 31 GB of space. We can also replace 7B with 13B, 30B, and 65B. The largest model takes up to 432 GB of space. We can use the below command to install llama model.

npx dalai llama install 7B

6. We can also select the model name and try various options.

References

https://github.com/ggerganov/llama.cpp

https://github.com/tatsu-lab/stanford_alpaca