Since github doesn't allow someone to write their issue as a more general one, I guess I'll choose the python repo (Since I will be using python anyways)

@Marcus-Peterson Sorry for the long delay. Thank you for raising this up. We quietly discontinued support for LLM training months ago, but the documentation didn't reflect it. This should now all be updated, and there's a public changelog entry to which the "Trainable language models" collection now redirects.