Run 01a03f0e · KernelIndex

Primary measurement

6.69ms±0.00 · mean of 100

Rank 1 in its comparison group · source-native comparison · observed 2026-03-05

max ms 6.71 · min ms 6.67 · std ms 0.00696 · mean ms 6.69

Identity

implementationtorch.compile (inductor)

projectPyTorch

revisionunknown

workloaddepth = 16 · width = 32 · height = 32 · batch_size = 128 · in_channels = 3 · fp32

comparison keysha256:62a763029bef1358…

sourceKernelBench baseline timings

external idH100_PCIe_LambdaLabs/torch-compile-inductor/level2/89_ConvTranspose3d_MaxPool_Softmax_Subtract_Swish_Max.py

sha256:1543023c89bbf72c7dafd11…

Correctness

Marked passed by the source; the correctness policy was not published.

Workload

depth16

width32

height32

batch_size128

in_channels3

input_1fp32 [128, 3, 16, 32, 32]

definition comparatornot_asserted

Measurements

latency · max6.71 ms · n=100

latency · mean6.69 ms · n=100

latency · min6.67 ms · n=100

latency · std6.96 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 manifest

{
  "run": {
    "kind": "BenchmarkRun",
    "spec": {
      "status": "passed",
      "timing": {
        "samples": 100,
        "latencyNs": {
          "mean": 6690000,
          "maximum": 6710000,
          "minimum": 6670000,
          "confidence95": [
            6688636,
            6691364
          ]
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2026-03-05T08:38:15.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 6960,
          "metric": "latency",
          "statistic": "std"
        }
      ],
      "sourceNative": {
        "source": "kernelbench",
        "metrics": {
          "max_ms": 6.71,
          "min_ms": 6.67,
          "std_ms": 0.00696,
          "mean_ms": 6.69
        },
        "benchmark": "H100_PCIe_LambdaLabs/baseline_time_torch_compile_inductor_default.json",
        "externalId": "H100_PCIe_LambdaLabs/torch-compile-inductor/level2/89_ConvTranspose3d_MaxPool_Softmax_Subtract_Swish_Max.py"
      },
      "protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
      "workloadDigest": "sha256:10d7d39194b79ed59cea3b53ddfe7878b4f352fe6bd43cce47fc56825fe7d44b",
      "environmentDigest": "sha256:89f9b28b06758f22a1df331fea58bfde2c9cc0f3627d1829509743ea538f0011",
      "implementationDigest": "sha256:333a117e37c5e495d67a9c1e5894ea5e1fc586e70d4c2f3905f7ea484ffba323"
    },
    "metadata": {
      "name": "kernelbench-h100-pcie-lambdalabs-torch-compile-inductor-l2-89-convtranspose3d-maxpool-softmax-subtra",
      "title": "ConvTranspose3d MaxPool Softmax Subtract Swish Max · torch.compile (inductor) · 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 ConvTranspose3d MaxPool Softmax Subtract Swish Max →JSON