[jramz]

That is cool, do you have a timeline set out for this?

[minimaxir]

I'll likely start by creating a web API service similar to what I did for gpt-2-simple, except more efficient: https://github.com/minimaxir/gpt-2-cloud-run

The next step is architecting an infrastructure for scalable generation; that depends on a few fixes for both aitextgen and the base Transformers. No ETA.