submission 128138
HayatoFujihara · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 16 lines, June 9 Researcher Reciprocity License v1.0.
submission.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-vectorsum-v2-128138?include=source"interfacepython
Compatibility
measured onNVIDIA A100
declared hardwareNVIDIA A100
architecturessm_80
dtypesfp32
Benchmark evidence
1 measurement across 1 GPU, fastest first.
Operation / workload
Hardware
Latency
Rank
Observed
Reported · How evidence levels are derived →
Source and license
sourceavailable
revision digestsha256:7cebabe10cbb8bc6a9766dc7052e1ce59a1777f7b57fb1b47e62bc81decdb449
license declaredunknown
license concludedunknown
authorsHayatoFujihara
imported2026-08-15
Kernel source
submission.py16 lines
import torch
from task import input_t, output_t
def custom_kernel(data: input_t) -> output_t:
"""
Fastest candidate: single-shot reduction.
- 全体を一発で float32 集約
- 出力は float32 スカラー
"""
input_tensor, _ = data
# 一発で float32 集約
result = torch.sum(input_tensor, dtype=torch.float32)
return result
Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0
Best evidence level for this revision: reported
JSON