In Operation
Fire up the software and you’ll see the home page which shows the apps we’ve installed.
The first you’ll want to do is click the Discover icon as that shows the AI apps you can install. It’s divided to two sections: Verified scripts and Community scripts.
The process of installing an app is simple. Pinokio uses scripts to install and configure various AI apps on your computer. Many of the apps can be deployed with a few clicks. Select an AI app from the list, click the download button, and the script takes care of everything needed to both install and configure the app. If any other software is needed on your system, you’ll be told.
Pinokio offers more than a simple way to install AI apps and models. It also lets you manage and run them, as well as the ability to create workflows to orchestrate installed AI apps. And there’s automation possibilities too, courtesy of an automation scripting language.
Everything is stored under an isolated folder and everything exists as a file, which means you can easily back up everything or delete apps simply by deleting files.
Summary
I give Pinokio a thumbs up. If you’re interested in dabbling with AI apps and your machine has a modern dedicated graphics card, it’s a good tool to use without having to learn about things like conda.
Many AI apps install out of the box with Pinokio. When Pinokio successfully installs an app it’s a real time-saver. But it’s far from perfect. For example, some AI apps refused to run on my system even confining myself to the verified scripts. For example, MeloTTS, PhotoMaker 2, and Hallucinator fail to run. In this situation, you’ll have to delve into the individual project’s install instructions. That’s a shame, as manually installing apps with things like virtual environments can be off putting for newcomers to Linux.
Pinokio’s Discover section offers a great range of AI apps, many of which I’d not heard of before. I’ll definitely review some of them and will include them in our Machine Learning in Linux series. A few of the apps are out of my range though as my dedicated graphics card 8GB VRAM isn’t enough for every app.
Pinokio is a self-contained system. You do not need to install any other program. Pinokio can automate anything, including program/library installations. Pinokio also offers many optimization features to save disk space.
Pinokio reminds me of the Arch User Repository (AUR) in a way. The AUR is a real time-saver for exploring open source apps with contributors submitting build scripts. It’s the main reason why I like Arch-based distros. In the same way as the AUR, contributors submit scripts for Pinokio. But it’s not uncommon for builds to fail in the AUR. It’s a mammoth task keeping build scripts working for tens of thousands of apps, so that’s not really surprising. And the AUR is full of detritus, software that relies on ancient dependencies or just abandoned software. Fortunately, Pinokio doesn’t suffer in that regard as it’s far more targeted.
Website: github.com/pinokiocomputer/pinokio
Support:
Developer: Pinokio
License: MIT License
Pinokio is written in JavaScript. Learn JavaScript with our recommended free books and free tutorials.
For other useful open source apps that use machine learning/deep learning, we’ve compiled this roundup.
Pages in this article:
Page 1 – Introduction and Installation
Page 2 – In Operation and Summary