pytorch/torch/backends/cuda
Eddie Yan 9ee506bd93 [CUDA][cuBLAS] Add fp16 accumulate option to cuBLAS/cuBLASLt (#144441)
Test for `cublasGemmEx` added, still need to figure out the best way to exercise the other APIs...

Pull Request resolved: https://github.com/pytorch/pytorch/pull/144441
Approved by: https://github.com/Chillee, https://github.com/malfet
2025-02-06 19:04:50 +00:00
..
__init__.py [CUDA][cuBLAS] Add fp16 accumulate option to cuBLAS/cuBLASLt (#144441) 2025-02-06 19:04:50 +00:00