Javascript is not enabled. This site can still works but it'll be more interactive when javascript is enabled.
loading...
Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
monster_truck
4 hours ago
|
on: DeepSeek V4 Flash on a Single AMD MI300X
Glazing over a lot, that's how they work already, just not in the way you think. A relatively small fraction of the model is active at any given time
reply