submission 780441
Kernel-Zhang · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 28 lines, June 9 Researcher Reciprocity License v1.0.
ref.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-matmul-v2-780441?include=source"interfacepython
Compatibility
measured onNVIDIA A100
declared hardwareNVIDIA A100
architecturessm_80
dtypesfp16
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:b83f48e6ea4749605c7a1480ba7559f33d9a6d79e3a9712b2009cff7db2aef0e
license declaredunknown
license concludedunknown
authorsKernel-Zhang
imported2026-08-15
Kernel source
ref.py28 lines
import torch
from task import input_t, output_t
from utils import make_match_reference, DeterministicContext
def custom_kernel(data: input_t) -> output_t:
a, b, c = data
return a @ b
def generate_input(m: int, n: int, k: int, seed: int) -> input_t:
gen = torch.Generator(device='cuda')
gen.manual_seed(seed)
a = torch.empty(m, k, device='cuda', dtype=torch.float16)
a.uniform_(0, 1, generator=gen)
b = torch.empty(k, n, device='cuda', dtype=torch.float16)
b.uniform_(0, 1, generator=gen)
c = torch.empty(m, n, device='cuda', dtype=torch.float16)
return a, b, c
def ref_kernel(data: input_t) -> output_t:
with DeterministicContext():
a, b, c = data
return a @ b
check_implementation = make_match_reference(ref_kernel)
Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0
Best evidence level for this revision: reported
JSON