So, my polyfill for Translator API works quite well. But I don’t know how to reasonably make up-to-date, high quality language models easily available for users, without requiring gigabytes of disk storage to host, say, pl-fe. I don’t want to bring too much centralization by hosting a model registry in one place for all pl-fe/polyfill users or favoring a single pl-fe deployment.

RT: https://pl.fediverse.pl/objects/b0908627-8d73-4a15-babc-7d803866a523

0

If you have a fediverse account, you can quote this note from your own instance. Search https://pl.fediverse.pl/objects/6606317e-ca96-4d08-bdee-9800861aa8ef on your instance and quote it. (Note that quoting is not supported in Mastodon.)