Hacker News new | past | comments | ask | show | jobs | submit
russli1993 | on: Kimi K3.1 Pro and K3.1 Pro Highspeed
since openAI and Anthropic are public about their anti-distillation mechanism, its either their anti-distillation doesn't work or kimi got here because their own training and RLs. Which one is it?
adenta | parent
I have no idea how you block distillation. Every time you do inference, thats basically distilation.