Alpaca-Turbo Homepage, Documentation and Downloads – Web UI for Running Alpaca Models Locally – News Delivery

Alpaca-Turbo is a language model that can run locally without much setup, it is based on the user-friendly web UI of LLaMA’s alpaca.cpp language model, the goal is to provide easy configuration and use without sacrificing speed or functionality seamless chat experience.

installation steps

1. Use Docker (docker only supports Linux)

Before doing this step,Docker must be installed on your system.

NOTE: For some reason this docker container works on linux but not windows

• frompublish pageDownload the latest alpaca-turbo.zip
• Extract the contents of the zip file into a directory called alpaca-turbo.
• put your Alpaca models are copied to the alpaca-turbo/models/ directory.
• Run the following commands to set everything up:

  docker-compose up

2. Windows/Mac M1/M2

• Install miniconda:
• frompublish pageDownload the latest alpaca-turbo.zip
• Extract Alpaca-Turbo.zip to Alpaca-Turbo(Make sure there is enough space for the model at the extracted location)
• put your Alpaca models are copied to the alpaca-turbo/models/ directory.
• Open cmd as administrator and enter

conda init

• close the window
• Open a new cmd window in the Alpaca-Turbo directory and enter

  conda create -n alpaca_turbo python=3.8 -y
  conda activate alpaca_turbo
  pip install -r requirements.txt
  python api.py

• Visit http://localhost:5000 , select your model and click Change, wait for the model to load

• ready to interact