Dataset advantage? #7
RichardScottOZ
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
It seems that natively Gemini (and Gemma) and amazon nova models (not tried others) are better at comics generally than Qwen, Mistral, etc.
Theory being that Google and Amazon (e.g. they literally took over Comixology) have an enormous amount of comics for sale so could easily be in the training data.
OpenAI models are terrible in comparison - if you ask Chat GPT 'are you trained in comics' - and it seems pretty much not. Google says it is.
Beta Was this translation helpful? Give feedback.
All reactions