submission 66761
kaide_max · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 7 lines, June 9 Researcher Reciprocity License v1.0.
vectorsum_beat1.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-vectorsum-v2-66761?include=source"interfacepython
Compatibility
measured onNVIDIA H100
declared hardwareNVIDIA H100
architecturessm_90
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 digestsha256:7e539304ab2b3b56488be7538181b33440f863a7819428e30594e02ed265e6ce
license declaredunknown
license concludedunknown
authorskaide_max
imported2026-08-15
Kernel source
vectorsum_beat1.py7 lines
import torch
def custom_kernel(input):
data = input[0] if isinstance(input, tuple) else input
data = data.contiguous() # Align for cache fuse
return data.sum(dim=0, dtype=torch.float32) # Explicit dim + float32—fused, skips ref double
Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0
Changes from previous submission
Against this author's previous submission submission 66756.
Best evidence level for this revision: reported
JSON