This project is a fork of the Original Project.
In this fork, I have introduced several enhancements to allow the customization of the server and model used by the application. The primary changes are as follows:
Three new environment variables have been added to the .env
file:
SERPER_API_KEY
: The serach API key for searching content online. You can also useBING_API_KEY
here.HELICONE_API_KEY
: The Helicone API key for observability.LLAMAEDGE_BASE_URL
: URL for the LLM API base URL.LLAMAEDGE_MODEL_NAME
: Name of the model to be used.LLAMAEDGE_API_KEY
: API key for accessing the LLM services.
These variables allow you to customize the URL and use your own server and model. The default values for these variables are:
LLAMAEDGE_BASE_URL=https://llama.us.gaianet.network/v1
LLAMAEDGE_MODEL_NAME=llama
LLAMAEDGE_API_KEY=LlamaEdge
git clone https://github.com/second-state/llamatutor.git
cd llamatutor
cp .example.env .env
Update the .env file with your desired values for the new variables:
LLAMAEDGE_BASE_URL=https://your-custom-url/v1
LLAMAEDGE_MODEL_NAME=your-custom-model
LLAMAEDGE_API_KEY=your-api-key
Apply for Serpre API key here and Helicone API key here.
npm install
or
yarn
npm run dev
or
yarn start
By configuring these environment variables, you can point the application to your own LLM server and model, providing greater flexibility and customization.