torch.compile (inductor)
conv transposed 3D asymmetric input asymmetric kernel strided padded grouped · depth = 12 · width = 48 · height = 24 · batch_size = 8 · in_channels = 32 · fp32 · run 01a03f0e-f6e2…
●passedReported evidence
Not re-observed recently.
Primary measurement
2.82ms±0.00 · mean of 100
Rank 1 in its comparison group · source-native comparison · observed 2026-03-05
max ms 2.86 · min ms 2.8 · std ms 0.00789 · mean ms 2.82
Identity
implementationtorch.compile (inductor)
projectPyTorch
revisionunknown
workloaddepth = 12 · width = 48 · height = 24 · batch_size = 8 · in_channels = 32 · fp32
comparison keysha256:2f195ccd7e01898b…
sourceKernelBench baseline timings
external idH100_Modal/torch-compile-inductor/level1/72_conv_transposed_3D_asymmetric_input_asymmetric_kernel___strided_padded_grouped_.py
sha256:d0ed740c55122f325e5a3de…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
depth12
width48
height24
batch_size8
in_channels32
xfp32 [8, 32, 12, 24, 48]
definition comparatornot_asserted
Measurements
latency · max2.86 ms · n=100
latency · mean2.82 ms · n=100
latency · min2.80 ms · n=100
latency · std7.89 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": 2820000,
"maximum": 2860000,
"minimum": 2800000,
"confidence95": [
2818454,
2821546
]
},
"primaryStatistic": "mean"
},
"observedAt": "2026-03-05T08:38:15.000Z",
"measurements": [
{
"unit": "ns",
"value": 7890,
"metric": "latency",
"statistic": "std"
}
],
"sourceNative": {
"source": "kernelbench",
"metrics": {
"max_ms": 2.86,
"min_ms": 2.8,
"std_ms": 0.00789,
"mean_ms": 2.82
},
"benchmark": "H100_Modal/baseline_time_torch_compile_inductor_default.json",
"externalId": "H100_Modal/torch-compile-inductor/level1/72_conv_transposed_3D_asymmetric_input_asymmetric_kernel___strided_padded_grouped_.py"
},
"protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
"workloadDigest": "sha256:23309f1abc74780b35ea1615313beae86b62a8aad00807f17fd50cdbcb738c1f",
"environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
"implementationDigest": "sha256:0176774fcba3c6aa28f1cd7ddd3aa6c82a286bb75333ccc8c5b9ef2662b5c9ff"
},
"metadata": {
"name": "kernelbench-h100-modal-torch-compile-inductor-l1-72-conv-transposed-3d-asymmetric-input-asymmetric-k",
"title": "conv transposed 3D asymmetric input asymmetric kernel strided padded grouped · torch.compile (inductor) · Modal"
},
"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": "80GB HBM3",
"architecture": "sm_90"
},
"software": {}
},
"metadata": {
"name": "kernelbench-h100-modal",
"title": "KernelBench host · H100 80GB HBM3 (Modal)"
},
"apiVersion": "kernelindex.dev/v1alpha1"
}
}Cite this record (permalink, digest, access date)