Introducing Triton: Open-source GPU programming for neural networks

Triton makes it possible to reach peak hardware performance with relatively little effort; for example, it can be used to write FP16 matrix multiplication kernels that match the performance of cuBLAS—something that many GPU programmers can’t do—in under 25 lines of code. Our researchers have already used it to produce kernels that are up to 2x more efficient than equivalent Torch implementations,

[Nyoho]

新言語だと……

[misshiki]

“オープンソースのPythonに似たプログラミング言語であるTriton1.0をリリース。CUDAの経験がない研究者でも、ほとんどの場合、専門家が作成できるものと同等の非常に効率的なGPUコードを記述できます。”