CUDA Kernels
topic on 2 shows · 4 statements across 4 episodes
4 statements about CUDA Kernels, every show
Socher: Recursive demonstrated its system can build new CUDA kernels
“We also showed that they can build new CUDA kernels, which is very useful for faster inference.”
Brockman: AI models will soon get very good at CUDA kernels
“Things like Cuda kernels are a good example of a very self-contained problem that actually our models should get very good at very soon, but it's just difficult because it requires a lot of domain expertise, a lot of like real abstract thinking. But again, it'…”
Tay discloses he does not know how to write CUDA kernels
“Should I know CUDA kernels? I don't know CUDA kernels.”
Chintala: Inference Moats from Fast CUDA Kernels Last Only Months
“I think, like, Together and Fireworks and all these people are trying to build some faster CUDA kernels and faster, like, you know, hardware kernels in general. But those modes only last for a month or two. Like, these ideas quickly propagate.”