Compare runs
cohort sha256:b4a2af03c52dfd25… · source-native profile
ranking-v1
no ranks: not comparable
Comparison identity
operation
kernelbench-l2-81-gemm-swish-divide-clamp-tanh-clamp
workload
batch_size = 1024 · in_features = 8192 · fp32 · 27ad84d0
protocol
KernelBench timing scripts · 547725b0
environment
NVIDIA H100 · 89f9b28b
correctness policy
52ff5e65
metric
latency mean (ns)
Context · 10 fields›
architecture
sm_90
CUDA
unknown
driver
unknown
framework
unknown
samples
100
evidence
reported
license
MIT
source
available
status
passed
observed
2026-03-05
No winner: these runs didn't measure the same thing. The rows above show what differs. Why comparable?
Add run ›
Export ›
Markdown CSV JSON
PyTorch eager · sha256:03284e5c38ced… · observed 2026-03-05