search

getumbrel / llama-gpt

A self-hosted, offline, ChatGPT-like chatbot. Powered by Llama 2. 100% private, with no data leaving your device. New: Code Llama support!
https://apps.umbrel.com/app/llama-gpt
MIT License
10.83k stars 700 forks source link

Update run.sh #166

Open true1ann opened 3 weeks ago

true1ann commented 3 weeks ago

The issue: original used docker compose instead of docker-compose, which lead to issues (docker compose does NOT support -f flag.)

other details:

BEFORE:

❯ sudo ./run.sh --model 7b --with-cuda
unknown shorthand flag: 'f' in -f
See 'docker --help'.

Usage:  docker [OPTIONS] COMMAND

A self-sufficient runtime for containers

Common Commands:
  run         Create and run a new container from an image
  exec        Execute a command in a running container
  ps          List containers
  ...

then i analyzed how run.sh was made, and added the change

AFTER:

❯ sudo ./run.sh --model 7b --with-cuda
WARN[0000] /home/<hidden>/projects/llamagpt/llama-gpt/docker-compose-cuda-ggml.yml: the attribute `version` is obsolete, it will be ignored, please remove it to avoid potential confusion
[+] Building 30.2s (4/28)
...

(i just made an install so it started compiling)