There are extreme diminishing returns in real world performance as models get bigger. 10x bigger might mean 5-10% better on benchmarks, a margin that can easily mean it's functionally equivalent in real world use or even a worse performer depending on the context it's being used in, and how good you are at providing meaningful context.
replyOf course the bigger model embeds more knowledge, but when neither model has the knowledge necessary to perform the task, hy3 makes idiotic decisions all the time whereas gemma 31b has a decent hit rate.
hy3 feels like someone who's read a lot of books and says the right words but has nothing of substance between their ears, gemma feels like a reasonably intelligent person who doesn't understand the domain, the latter is muuuch easier to work with than the former.
Gemma 4 is the first really small model that feels smart, to me. I mean, Qwen 3.6 is arguably better at some coding tasks. But, Gemma 4 has shockingly good reasoning for a small model. Even the tiny 12B, at 7GB on disk in the 4-bit QAT quant, feels like a really big model of a couple years ago. It's a good tool user, can search the web (when given the appropriate skill or MCP), has good vision capabilities, and pretty good prose.
replyI've only used the Hy3 preview, so I don't want to judge too harshly, yet. But, I wasn't very impressed with it a couple of months ago.