torch.compile (inductor)
ConvTranspose3d Clamp Min Divide · depth = 24 · width = 48 · height = 48 · batch_size = 16 · in_channels = 64 · fp32 · run 01a03f0e-f6ed…
●passedReported evidence
Not re-observed recently.
Primary measurement
8.29ms±0.04 · mean of 100
Rank 1 in its comparison group · source-native comparison · observed 2026-03-05
max ms 10.5 · min ms 8.23 · std ms 0.224 · mean ms 8.29
Identity
implementationtorch.compile (inductor)
projectPyTorch
revisionunknown
workloaddepth = 24 · width = 48 · height = 48 · batch_size = 16 · in_channels = 64 · fp32
comparison keysha256:5ca9a310bcd32ad1…
sourceKernelBench baseline timings
external idH100_Modal/torch-compile-inductor/level2/100_ConvTranspose3d_Clamp_Min_Divide.py
sha256:c8696f5b7f5b13c12f3a161…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
depth24
width48
height48
batch_size16
in_channels64
input_1fp32 [16, 64, 24, 48, 48]
definition comparatornot_asserted
Measurements
latency · max10.5 ms · n=100
latency · mean8.29 ms · n=100
latency · min8.23 ms · n=100
latency · std224.0 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": 8290000,
"maximum": 10500000,
"minimum": 8230000,
"confidence95": [
8246096,
8333904
]
},
"primaryStatistic": "mean"
},
"observedAt": "2026-03-05T08:38:15.000Z",
"measurements": [
{
"unit": "ns",
"value": 224000,
"metric": "latency",
"statistic": "std"
}
],
"sourceNative": {
"source": "kernelbench",
"metrics": {
"max_ms": 10.5,
"min_ms": 8.23,
"std_ms": 0.224,
"mean_ms": 8.29
},
"benchmark": "H100_Modal/baseline_time_torch_compile_inductor_default.json",
"externalId": "H100_Modal/torch-compile-inductor/level2/100_ConvTranspose3d_Clamp_Min_Divide.py"
},
"protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
"workloadDigest": "sha256:25c761cb439d2d600e661e0d1daa808f09a7c75c797680d59b01ca1f86db1cda",
"environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
"implementationDigest": "sha256:69d8510f3d7dfe8101ca0aff5144fd9eacec11c0f5ceb3a7bac04ad9ecd8ab96"
},
"metadata": {
"name": "kernelbench-h100-modal-torch-compile-inductor-l2-100-convtranspose3d-clamp-min-divide",
"title": "ConvTranspose3d Clamp Min Divide · 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)
Report an issue with this run
Published 2026-08-26 · KernelBench baseline timings · MITAll results for ConvTranspose3d Clamp Min Divide →JSON