Tabby - Free And Opensource Copilot Alternative

With Code Example
April 1, 2024

Tabby is a self-hosted AI coding assistant, offering an open-source and on-premises alternative to GitHub Copilot. It boasts several key features:

Hello Developers, If you looking free alternative to GitHub Copilot (AI coding assistant), then you are in the right place. In recent days I got to know about one awesome open-source self-hosted tool for AI coding assistant tool, called Tabby.

screenshot-2024-04-01-at-4.53.10 pm_DX5iX5hh_8907324567068119886.jpg

Table of Contents

According to their documentation, tabby has the following features

Tabby is a self-hosted AI coding assistant that provides an alternative to GitHub Copilot, which is open-source and on-premises.

First of all, it is self-contained. This means that users do not need a DBMS or cloud service.
Secondly, it has an OpenAPI interface. Consequently, the language makes it easy for them to integrate with existing infrastructure (for example Cloud IDE).
Thirdly, Tabby supports consumer-grade GPUs.

tabby

Install

You can install Tabby through various methods such as docker, docker-compose, Homebrew(Mac) and Windows release. Let’s check them one by one.

Install Tabby with Docker

With GPU (NVIDIA)
CPU

docker run -it --gpus all \
  -p 8080:8080 -v $HOME/.tabby:/data \
  tabbyml/tabby serve --model StarCoder-1B --device cuda

docker run --entrypoint /opt/tabby/bin/tabby-cpu -it \
  -p 8080:8080 -v $HOME/.tabby:/data \
  tabbyml/tabby serve --model StarCoder-1B

Install Tabby with docker-compose file

With GPU (NVIDIA)
CPU

version: '3.5'

services:
  tabby:
    restart: always
    image: tabbyml/tabby
    command: serve --model StarCoder-1B --device cuda
    volumes:
      - "$HOME/.tabby:/data"
    ports:
      - 8080:8080
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

version: '3.5'

services:
  tabby:
    restart: always
    image: tabbyml/tabby
    entrypoint: /opt/tabby/bin/tabby-cpu
    command: serve --model StarCoder-1B
    volumes:
      - "$HOME/.tabby:/data"
    ports:
      - 8080:8080

Install Tabby with Homebrew (Apple M1/M2)

# install
brew install tabbyml/tabby/tabby

# Start server with StarCoder-1B
tabby serve --device metal --model StarCoder-1B

Install Tabby with Windows Release

Go to the Tabby release page: https://github.com/TabbyML/tabby/releases
Click on the Assets dropdown for a specific release to find the Windows executable (exe) files and download it.
running tabby on windows .\tabby_x86_64-windows-msvc-cuda117.exe serve --model StarCoder-1B --device cuda