Y
Hacker News
new
|
ask
|
show
|
jobs
by
HarHarVeryFunny
168 days ago
The entire history of RL-trained "reasoning models" from o1 to DeepSeek_R1 is basically just a year old!