Skip to content
KernelIndex
Search⌘K

Transformers

GroupNorm · batch = 128 · hidden = 512 · channels = 128 · channels_per_group = 4 · fp32 · run 01a0202b-2351…
●passedReported evidence

Not re-observed recently.

Primary measurement
133.2µsmedian

Rank 2 in its comparison group · source-native comparison · observed 2026-02-28

Compare with #1 →

ms 20 0.13308799266815186 · ms 50 0.13315199315547943 · ms 80 0.13331200182437897

Identity
implementationTransformers
projectHugging Face Transformers
revisionunknown
workloadbatch = 128 · hidden = 512 · channels = 128 · channels_per_group = 4 · fp32
comparison keysha256:acc3f8fcc477edac…
sourceLiger-Kernel benchmarks
external idgroup_norm/backward/huggingface/nvidia-b200/batch128-channels128-channels_per_group4-hidden512-fp32/2026-02-28-00-23-31
sha256:5fcedc12131d049cf97bff2…

Correctness

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

Workload

batch128
hidden512
channels128
channels_per_group4
xfp32 [128, 128, 512]
biasfp32 [128]
weightfp32 [128]
definition comparatornot_asserted

Measurements

latency · median133.2 µs
latency · p20133.1 µs
latency · p80133.3 µs

Protocol

harnessLiger-Kernel benchmark scripts
timercuda_events
primaryStatisticmedian
comparabilityFamilyliger_kernel_bench

Environment

gpuNVIDIA B200 (sm_100)

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
I

Published under your account name. Community attestations never change the evidence level; a maintainer can hide abuse.

Canonical manifest

Show manifest
{
  "run": {
    "kind": "BenchmarkRun",
    "spec": {
      "status": "passed",
      "timing": {
        "latencyNs": {
          "median": 133152
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2026-02-28T00:23:31.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 133088,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 133312,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 0.13308799266815186,
          "ms_50": 0.13315199315547943,
          "ms_80": 0.13331200182437897
        },
        "benchmark": "group_norm/backward",
        "externalId": "group_norm/backward/huggingface/nvidia-b200/batch128-channels128-channels_per_group4-hidden512-fp32/2026-02-28-00-23-31"
      },
      "protocolDigest": "sha256:080e6e66e4ec75a131aa0c9121c74b8e8eed64e7b6d621d567c6a28f79548be8",
      "workloadDigest": "sha256:d36b540755870755faf627039214befc9ad8dae4e6e66baf8e9ef3545a5a0007",
      "environmentDigest": "sha256:ad459e643ef802f7008f1fa8915785c905d05abbfdb5eb0328be214f8d9f3a8b",
      "implementationDigest": "sha256:e8c7d555b6002524eb04f3e8a91c03603172925143b51877f4050be074e9e025"
    },
    "metadata": {
      "name": "liger-group-norm-backward-huggingface-nvidia-b200-batch128-channels128-channels-per-group4-hidden512",
      "title": "GroupNorm · huggingface · backward",
      "labels": {
        "liger_version": "0.7.0"
      }
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "protocol": {
    "kind": "BenchmarkProtocol",
    "spec": {
      "harness": {
        "name": "Liger-Kernel benchmark scripts",
        "repository": "https://github.com/linkedin/Liger-Kernel"
      },
      "measurement": {
        "timer": "cuda_events",
        "primaryStatistic": "median"
      },
      "comparability": {
        "notes": "Timed: the backward pass of the module output given a random upstream gradient, via triton.testing.do_bench with quantiles 0.5/0.2/0.8 (median with p20/p80 spread). The upstream CSV records no CUDA, driver, or torch version; the Liger release rides each run's labels. Comparable only within one kernel, workload, timed pass, and GPU.",
        "family": "liger_kernel_bench"
      }
    },
    "metadata": {
      "name": "liger-bench-do-bench-backward",
      "title": "Liger benchmark harness · backward pass"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "environment": {
    "kind": "ExecutionEnvironment",
    "spec": {
      "hardware": {
        "vendor": "nvidia",
        "product": "NVIDIA B200",
        "architecture": "sm_100"
      },
      "software": {}
    },
    "metadata": {
      "name": "liger-bench-nvidia-b200",
      "title": "Liger benchmark host · B200"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}
Cite this record (permalink, digest, access date)
Report an issue with this run

A maintainer reviews every report. If accepted, the record is retracted or superseded; its history stays visible.

Published 2026-08-20 · Liger-Kernel benchmarks · BSD-2-ClauseAll results for GroupNorm →JSON