Open Assistant, the launch of a ‘ChatGPT replicant’
Newly launched and open source, Open Assistant represents the arrival of a ‘ChatGPT replicant’ for the user/developer community.
The project’s intention is to create a chat-based large language model.
Detailing the software’s core mechanics on the project’s GitHub page, the team is hoping to create what it calls ‘a revolution in innovation in language’ (by which they mean human-computer language interaction experiences… and, in human language itself) in the same way that Stable Diffusion worked to enable people to work with art and images.
NOTE: Stable Diffusion is a deep learning, text-to-image model released in 2022.
Open Assistant was developed under LAION: Large-Scale Artificial Intelligence Open Network
NOTE: LAION provides datasets, tools and models to liberate machine learning research. By doing so, LAION says it encourages open public education and a more environmentally-friendly use of resources by reusing existing datasets and models.
Open Assistant is primarily built in Python and TypeScript, with a portion of JavaScript as well.
According to Open Assistant, “We are not going to stop at replicating ChatGPT. We want to build the assistant of the future, able to not only write email and cover letters, but do meaningful work, use APIs, dynamically research information, and much more, with the ability to be personalized and extended by anyone. And we want to do this in a way that is open and accessible, which means we must not only build a great assistant, but also make it small and efficient enough to run on consumer hardware.
The team is calling for contributors and collaborators as of now.
Get involved
They confirm that there is ‘a growing task list of issues’, so developers and interested parties should find an issue that appeals and make a comment to work on it.
“Include in your comment a brief description of how you’ll solve the problem and if there are any open questions you want to discuss. Once a project coordinator has assigned the issue to you, start working on it,” notes the team.
Upon making a release on GitHub, all docker images are automatically built and pushed to ghcr.io.