Y
Hacker News
new
|
ask
|
show
|
jobs
by
apitman
1 day ago
Are you sure the difference is from MoE and not that 3.6 is newer?
1 comments
riknos314
1 day ago
Qwen 3.5 27B also scores higher than 3.5 122BA10B. So even in the same generation the smaller dense model outperformed the larger MOE
link
riknos314
1 day ago
https://artificialanalysis.ai/evaluations/artificial-analysi...
link
apitman
23 hours ago
Ah gotcha, I hadn't noticed that. Thanks
link