submission 396744
gau.nernst · python · License unknown
Use it
Vendorable · source mirrored · license unknownView source →
No package. Vendor the mirrored source: 20 lines, June 9 Researcher Reciprocity License v1.0.
submission_ref.py
curl "https://kernelindex.com/api/v1/implementations/kernelbot-nvfp4-group-gemm-396744?include=source"interfacepython
Compatibility
measured onNVIDIA B200
declared hardwareNVIDIA B200
architecturessm_100
dtypesfp8_e4m3, nvfp4
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:f43feb5c6d116d6c919ab564cd9cd4a2b1682f6dbc300d15c85bedd963515a86
license declaredunknown
license concludedunknown
authorsgau.nernst
imported2026-08-15
Kernel source
submission_ref.py20 lines
#!POPCORN leaderboard nvfp4_group_gemm
#!POPCORN gpu B200
import torch
from task import input_t, output_t
def custom_kernel(data: input_t) -> output_t:
abc_list, _, sf_list, _ = data
return [
torch._scaled_mm(
a[..., 0],
b[..., 0].T,
sfa.permute(5, 2, 4, 0, 1, 3).view(-1),
sfb.permute(5, 2, 4, 0, 1, 3).view(-1),
out_dtype=torch.float16,
).unsqueeze(-1)
for (a, b, _), (sfa, sfb) in zip(abc_list, sf_list)
]
Source code from GPU Mode and the KernelBot dataset · June 9 Researcher Reciprocity License v1.0
Best evidence level for this revision: reported
JSON