Open-Assistant/docs/docs/intro.md

Introduction

OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

It can be extended and personalized easily and is developed as free, open-source software.

Our Vision

We want OpenAssistant to be the single, unifying platform that all other systems use to interface with humans.

Principles

• We put the human in the center
• We need to get the MVP out fast, while we still have momentum
• We pull in one direction
• We are pragmatic
• We aim for models that can (or could, with some effort) be run on consumer hardware
• We rapidly validate our ML experiments on a small scale, before going to a supercluster

Main Efforts

• Data Collection Code → Backend, website, and discord bot to collect data
• Instruction Dataset Gathering → Scraping & cleaning web data
• Gamification → Leaderboards & more, to make data collection more fun
• Model Training → Experiments on pseudo- and real-data
• Infrastructure → Collection, training, and inference
• Data Collection → This is the bulk of the work
• Data Augmentation → Making more data from little data
• Privacy and Safety → Protecting sensitive data