Skip to content
KernelIndex
Search⌘K

submission 512880

JordanNanos · python · License unknown

Use it

Vendorable · source mirrored · license unknownView source →

No package. Vendor the mirrored source: 13 lines, June 9 Researcher Reciprocity License v1.0.

prefixsum_py_submission.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-prefixsum-v2-512880?include=source"
interfacepython
Compatibility
measured onNVIDIA B200
declared hardwareNVIDIA B200
architecturessm_100
dtypesfp32

Benchmark evidence

1 measurement across 1 GPU, fastest first.

Operation / workload
Hardware
Latency
Rank
Observed
Inclusive prefix sumsuite of 11 cases
NVIDIA B200
3.27ms
#21 of 23
2026-03-05

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:6f5b4de9e6326044228d182d3df1a995d7db3421aefad62f83818dbeb255e240
license declaredunknown
license concludedunknown
authorsJordanNanos
imported2026-08-15

Kernel source

prefixsum_py_submission.py13 lines
import torch
from task import input_t, output_t


def custom_kernel(data: input_t) -> output_t:
    x, output = data
    n = x.numel()
    if n == 0:
        return output
    result = torch.cumsum(x.to(torch.float64), dim=0).to(torch.float32)
    output.copy_(result)
    return output

Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0

Best evidence level for this revision: reported

JSON