The tokenizer changes seem to indicate that 4.7 isn't just a checkpoint but rath... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		oezi 40 days ago \| parent \| context \| favorite \| on: Claude Opus 4.7 The tokenizer changes seem to indicate that 4.7 isn't just a checkpoint but rather a model trained mostly from scratch, right?

dannyw 40 days ago [–]

You can change tokenizers without a complete retraining from scratch.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact