Thursday, August 1, 2024

Show HN: FP32 matmul of large matrices up to 24% faster than cuBLAS on a 4090 https://ift.tt/brZGm0U

Show HN: FP32 matmul of large matrices up to 24% faster than cuBLAS on a 4090 I decided to share a CUDA kernel I wrote over 5 months ago. Nvidia's hardware and software may surprise you. https://ift.tt/HrKBPtx August 1, 2024 at 12:09AM
-
Labels:

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)

Show HN: Pocket2Linkding – Migrate from Mozilla Pocket to Linkding https://ift.tt/IwYJfju

Show HN: Pocket2Linkding – Migrate from Mozilla Pocket to Linkding With the Mozilla Pocket shutdown coming up in about two weeks, I thought ...