It kind of blows me away how “unreasonably effective” these small image models seem for their size.
Meanwhile a decent LLM that can do tool use and code and reason tends to weigh in at a minimum of 30B parameters, and you need over 100B to get good performance.
Do LLMs just contain more information or are diffusion models more efficient at compression?
I saw some work on language diffusion models a while back but I think they tend to have problems with reasoning versus transformer type models.
Meanwhile a decent LLM that can do tool use and code and reason tends to weigh in at a minimum of 30B parameters, and you need over 100B to get good performance.
Do LLMs just contain more information or are diffusion models more efficient at compression?
I saw some work on language diffusion models a while back but I think they tend to have problems with reasoning versus transformer type models.