Improve Fused8BitRowwiseQuantizedSBFloatToFloatOrHalfNeon by 5%-15% · pytorch/FBGEMM@254396f

Triggered via pull request March 21, 2025 15:07

Nicoshev

opened #3860

Status Success

Total duration 38m 4s

Artifacts 2

fbgemm_gpu_benchmark_cpu.yml

on: pull_request

Matrix: build_artifact

Matrix: benchmark_artifact

Produced during runtime

Name	Size	Digest
fbgemm_gpu_nightly_cpu_arm_gcc_py3.13.whl	3.76 MB	`sha256:9aefb54ca5fb19c1c83233fd49f906ed0c2b3038d2f6a9eba591f12694a48971`
fbgemm_gpu_nightly_cpu_x86_gcc_py3.13.whl	4.91 MB	`sha256:279874a9faf58ebeef413b357a5840941153c47581ccb3e950e2f394c360c9ed`