Skip to content
KernelIndex
Search⌘K

submission 866036

.satan_99 · python · License unknown

Use it

Vendorable · source mirrored · license unknownView source →

No package. Vendor the mirrored source: 11 lines, June 9 Researcher Reciprocity License v1.0.

submission_v3.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-eigh-866036?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
NVIDIA B200
48.9ms
#153 of 286
2026-07-10

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:f25822194cf65f5181454836ffe348a461f771a3d3353ed6469bc28e3fc4986c
license declaredunknown
license concludedunknown
authors.satan_99
imported2026-08-26

Kernel source

submission_v3.py11 lines
import torch
from task import input_t, output_t

torch.backends.cuda.preferred_linalg_library("cusolver")


@torch.inference_mode()
def custom_kernel(data: input_t) -> output_t:
    values, vectors = torch.linalg.eigh(data, UPLO="L")
    return vectors, values

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

Best evidence level for this revision: reported

JSON