Hacker News new | ask | show | jobs
by floodfx 216 days ago
Why are completion tokens more with image prompts yet the text output was about the same?
2 comments
cma 215 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.
link
Garlef 216 days ago
"Thinking" Mode
link
nunodonato 215 days ago
it doesn't say that anywhere.
link