Run 01a0202b · KernelIndex

Primary measurement

21.5msmedian

Rank 2 in its comparison group · source-native comparison · observed 2026-04-02

Compare with #1 →

ms 20 21.456192016601562 · ms 50 21.456192016601562 · ms 80 21.456192016601562

Identity

implementationTransformers

projectHugging Face Transformers

revisionunknown

workloadk = 8 · hidden = 2048 · tokens = 8192 · experts = 32 · intermediate = 768 · bf16

comparison keysha256:7b72896ede4199ee…

sourceLiger-Kernel benchmarks

external idfused_moe/backward/huggingface/nvidia-h100-80gb-hbm3/experts32-hidden2048-intermediate768-k8-tokens8192-bf16/2026-04-02-23-59-47

sha256:c546af37bd0bfc7b7098795…

Correctness

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

Workload

hidden2048

tokens8192

experts32

intermediate768

xbf16 [8192, 2048]

down_projbf16 [32, 2048, 768]

top_k_indexint32 [8192, 8]

gate_up_projbf16 [32, 1536, 2048]

top_k_weightsbf16 [8192, 8]

definition comparatornot_asserted

Measurements

latency · median21.5 ms

latency · p2021.5 ms

latency · p8021.5 ms

Protocol

harnessLiger-Kernel benchmark scripts

timercuda_events

primaryStatisticmedian

comparabilityFamilyliger_kernel_bench

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": {
        "latencyNs": {
          "median": 21456192
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2026-04-02T23:59:47.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 21456192,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 21456192,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 21.456192016601562,
          "ms_50": 21.456192016601562,
          "ms_80": 21.456192016601562
        },
        "benchmark": "fused_moe/backward",
        "externalId": "fused_moe/backward/huggingface/nvidia-h100-80gb-hbm3/experts32-hidden2048-intermediate768-k8-tokens8192-bf16/2026-04-02-23-59-47"
      },
      "protocolDigest": "sha256:080e6e66e4ec75a131aa0c9121c74b8e8eed64e7b6d621d567c6a28f79548be8",
      "workloadDigest": "sha256:a5a07bf085162e8fa6408b04ae2f483b4b98084d43cde827d7e308042d07f376",
      "environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
      "implementationDigest": "sha256:205272a6edab93cf893ad6d2fa2cc9786bfbfa97a1a7fdab0ed7f41768333593"
    },
    "metadata": {
      "name": "liger-fused-moe-backward-huggingface-nvidia-h100-80gb-hbm3-experts32-hidden2048-intermediate768-k8-t",
      "title": "Fused MoE expert MLP · 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 H100",
        "formFactor": "80GB HBM3",
        "architecture": "sm_90"
      },
      "software": {}
    },
    "metadata": {
      "name": "liger-bench-nvidia-h100-80gb-hbm3",
      "title": "Liger benchmark host · H100 80GB HBM3"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}

Cite this record (permalink, digest, access date)

Report an issue with this run

Published 2026-08-20 · Liger-Kernel benchmarks · BSD-2-ClauseAll results for Fused MoE expert MLP →JSON