PyTorch eager
Max Pooling 2D · width = 512 · height = 512 · channels = 64 · batch_size = 32 · fp32 · run 01a03f0e-f832…
●passedReported evidence
Not re-observed recently.
Primary measurement
14.1ms±0.00 · mean of 100
Rank 2 in its comparison group · source-native comparison · observed 2026-03-05
max ms 14.1 · min ms 14.1 · std ms 0.00888 · mean ms 14.1
Identity
implementationPyTorch eager
projectPyTorch
revisionunknown
workloadwidth = 512 · height = 512 · channels = 64 · batch_size = 32 · fp32
comparison keysha256:f133cb9d077326f9…
sourceKernelBench baseline timings
external idH100_PCIe_LambdaLabs/torch/level1/42_Max_Pooling_2D.py
sha256:e7f8d93d08d74baf970138d…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
width512
height512
channels64
batch_size32
xfp32 [32, 64, 512, 512]
definition comparatornot_asserted
Measurements
latency · max14.1 ms · n=100
latency · mean14.1 ms · n=100
latency · min14.1 ms · n=100
latency · std8.88 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": 14100000,
"maximum": 14100000,
"minimum": 14100000,
"confidence95": [
14100000,
14101740
]
},
"primaryStatistic": "mean"
},
"observedAt": "2026-03-05T08:38:15.000Z",
"measurements": [
{
"unit": "ns",
"value": 8880,
"metric": "latency",
"statistic": "std"
}
],
"sourceNative": {
"source": "kernelbench",
"metrics": {
"max_ms": 14.1,
"min_ms": 14.1,
"std_ms": 0.00888,
"mean_ms": 14.1
},
"benchmark": "H100_PCIe_LambdaLabs/baseline_time_torch.json",
"externalId": "H100_PCIe_LambdaLabs/torch/level1/42_Max_Pooling_2D.py"
},
"protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
"workloadDigest": "sha256:a35a5828e74e09e31c2bdfa7ba34ecd3a41fe0477d8e93e9698bc4f93dab5603",
"environmentDigest": "sha256:89f9b28b06758f22a1df331fea58bfde2c9cc0f3627d1829509743ea538f0011",
"implementationDigest": "sha256:fbabafe242fb72d358db091983fc46d45698fb27fd89542ed16880e9fdfb7f61"
},
"metadata": {
"name": "kernelbench-h100-pcie-lambdalabs-torch-l1-42-max-pooling-2d",
"title": "Max Pooling 2D · 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 Max Pooling 2D →JSON