submission 553516
canofspam. · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 18 lines, June 9 Researcher Reciprocity License v1.0.
submission.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-grayscale-v2-553516?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
Reported · How evidence levels are derived →
Source and license
sourceavailable
revision digestsha256:03c9ad7d9d20594cb04d29a214ee170321119e3450fef58e418ba2ce71696b45
license declaredunknown
license concludedunknown
authorscanofspam.
imported2026-08-15
Kernel source
submission.py18 lines
#!POPCORN leaderboard grayscale_v2
#!POPCORN gpu B200
from task import input_t, output_t
import torch
@torch.compile
def compiled_matmul(a, b):
return a @ b
def custom_kernel(data: input_t) -> output_t:
data, output = data
weights = torch.tensor([0.2989, 0.5870, 0.1140],
device=data.device,
dtype=data.dtype)
output[...] = compiled_matmul(data, weights)
return output
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 552488.
+ #!POPCORN leaderboard grayscale_v2+ #!POPCORN gpu B200+from task import input_t, output_timport torch+ @torch.compile+ def compiled_matmul(a, b):+ return a @ b+def custom_kernel(data: input_t) -> output_t:data, output = dataweights = torch.tensor([0.2989, 0.5870, 0.1140],device=data.device,dtype=data.dtype)- output[...] = torch.sum(data * weights, dim=-1)+ output[...] = compiled_matmul(data, weights)return output
Best evidence level for this revision: reported
JSON