RadixArk
@radixark
We're proud that @periodiclabs chose SGLang and Miles to build Neon.
Periodic extended SGLang and Miles to run scientific RL at trillion-parameter scale, with more efficient training, lower memory use, and 2.5x faster inference. This work was contributed back to both projects.
Periodic extended SGLang and Miles to run scientific RL at trillion-parameter scale, with more efficient training, lower memory use, and 2.5x faster inference. This work was contributed back to both projects.
Liam Fedus@LiamFedus · Sep 15We built high-throughput materials labs in Menlo Park to create a loop between experiments and models. The labs generate fresh data, the models learn from it, and then help us decide what to try next.
Using only 1,300 H200s, plus months of our experimental data, we mid-trained
Using only 1,300 H200s, plus months of our experimental data, we mid-trained
0:16 1 113