Skip to content

deploy.md

Latest commit

 

History

History
31 lines (23 loc) · 993 Bytes

deploy.md

File metadata and controls

31 lines (23 loc) · 993 Bytes

How to deploy the app

Prerequisites

  1. Download gguf weights from Hugging Face
bash llama-cpp-server/download_gguf.sh

Laucnh the app

  1. create a .env.prod file (or whatever the name is) in the root directory of the project and add the following environment variables:
LINE_CHANNEL_ACCESS_TOKEN=your_line_channel_access_token
LINE_CHANNEL_SECRET=your_line_channel_secret
NGROK_AUTHTOKEN=your_ngrok_authtoken
HF_API_TOKEN_LIST=your_huggingface_api_token_1 your_huggingface_api_token_2
  1. run the following command to start the app:
docker compose --env-file=.env.config --env-file=.env.prod up -d && sleep 1 && docker compose logs ngrok -n all | grep url=https:
  1. copy the ngrok url and set the webhook url in your LINE developer console to https://your_ngrok_url/callback. (make sure to append /callback to the ngrok url)

How do inspect logs?

docker compose logs -f