Why are completion tokens more with image prompts yet the text output was about ... | Hacker News

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

		floodfx 44 days ago \| parent \| context \| favorite \| on: Can you save on LLM tokens using images instead of... Why are completion tokens more with image prompts yet the text output was about the same?

cma 44 days ago | [–]

Some multimodal models may have a hidden captioning step that may take completion tokens, others work on a fully native representation, and some do both I think.

Garlef 44 days ago | [–]

"Thinking" Mode

nunodonato 44 days ago | [–]

it doesn't say that anywhere.

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