PyTorch eager
Conv2d Tanh Scaling BiasAdd Max · width = 256 · height = 256 · batch_size = 128 · in_channels = 8 · fp32 · run 01a03f0e-f83d…
●passedReported evidence
Not re-observed recently.
Primary measurement
14.7ms±0.05 · mean of 100
Rank 2 in its comparison group · source-native comparison · observed 2026-03-05
max ms 15 · min ms 14.4 · std ms 0.244 · mean ms 14.7
Identity
implementationPyTorch eager
projectPyTorch
revisionunknown
workloadwidth = 256 · height = 256 · batch_size = 128 · in_channels = 8 · fp32
comparison keysha256:4ae0fa464cdaccba…
sourceKernelBench baseline timings
external idH100_PCIe_LambdaLabs/torch/level2/82_Conv2d_Tanh_Scaling_BiasAdd_Max.py
sha256:95463a21498d1061ebb8089…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
width256
height256
batch_size128
in_channels8
input_1fp32 [128, 8, 256, 256]
definition comparatornot_asserted
Measurements
latency · max15.0 ms · n=100
latency · mean14.7 ms · n=100
latency · min14.4 ms · n=100
latency · std244.0 µs
Protocol
harnessKernelBench timing scripts
timercuda_events
primaryStatisticmean
comparabilityFamilykernelbench_baseline_timing
Environment
gpuNVIDIA H100 (sm_90)
Artifacts
No artifacts published with this run.
Replications and notes
No attestations yet.
Community attestations. They never change the evidence level; only a KernelIndex-controlled rerun does.
Add a reproduction or note
Canonical manifest
Show manifestHide manifest
{
"run": {
"kind": "BenchmarkRun",
"spec": {
"status": "passed",
"timing": {
"samples": 100,
"latencyNs": {
"mean": 14700000,
"maximum": 15000000,
"minimum": 14400000,
"confidence95": [
14652176,
14747824
]
},
"primaryStatistic": "mean"
},
"observedAt": "2026-03-05T08:38:15.000Z",
"measurements": [
{
"unit": "ns",
"value": 244000,
"metric": "latency",
"statistic": "std"
}
],
"sourceNative": {
"source": "kernelbench",
"metrics": {
"max_ms": 15,
"min_ms": 14.4,
"std_ms": 0.244,
"mean_ms": 14.7
},
"benchmark": "H100_PCIe_LambdaLabs/baseline_time_torch.json",
"externalId": "H100_PCIe_LambdaLabs/torch/level2/82_Conv2d_Tanh_Scaling_BiasAdd_Max.py"
},
"protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
"workloadDigest": "sha256:70533588eaace09760fc3b5b0667bd95ff54b2b336516b3fc7fd15de27ed09cb",
"environmentDigest": "sha256:89f9b28b06758f22a1df331fea58bfde2c9cc0f3627d1829509743ea538f0011",
"implementationDigest": "sha256:efecbca5c5a2da94e157433ceb233c45d1ff90b3f8b2c8e5df6060ddd2385708"
},
"metadata": {
"name": "kernelbench-h100-pcie-lambdalabs-torch-l2-82-conv2d-tanh-scaling-biasadd-max",
"title": "Conv2d Tanh Scaling BiasAdd Max · PyTorch eager · Lambda Labs"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"protocol": {
"kind": "BenchmarkProtocol",
"spec": {
"harness": {
"name": "KernelBench timing scripts",
"repository": "https://github.com/ScalingIntelligence/KernelBench"
},
"measurement": {
"timer": "cuda_events",
"primaryStatistic": "mean"
},
"comparability": {
"notes": "Mean of 100 timed forward passes (CUDA events, warm-up excluded) of the reference module on fixed inputs; the 95% interval is the mean's, from the reported standard deviation. The upstream JSON records no torch or CUDA version. Comparable only within one problem and one timing host.",
"family": "kernelbench_baseline_timing"
}
},
"metadata": {
"name": "kernelbench-timing-v1",
"title": "KernelBench baseline timing"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"environment": {
"kind": "ExecutionEnvironment",
"spec": {
"hardware": {
"vendor": "nvidia",
"product": "NVIDIA H100",
"formFactor": "PCIe",
"architecture": "sm_90"
},
"software": {}
},
"metadata": {
"name": "kernelbench-h100-pcie-lambdalabs",
"title": "KernelBench host · H100 PCIe (Lambda Labs)"
},
"apiVersion": "kernelindex.dev/v1alpha1"
}
}Cite this record (permalink, digest, access date)
Report an issue with this run
Published 2026-08-26 · KernelBench baseline timings · MITAll results for Conv2d Tanh Scaling BiasAdd Max →JSON