Run 01a0202b · KernelIndex

Primary measurement

20.9msmedian

Rank 2 in its comparison group · source-native comparison · observed 2025-05-27

Compare with #1 →

ms 20 20.860185623168945 · ms 50 20.872079849243164 · ms 80 21.320632934570312

Identity

implementationPyTorch

projectPyTorch

revisionunknown

workloadseq = 256 · batch = 4 · heads = 12 · hidden = 768 · dilation = 1 · kernel_size = 7 · fp32

comparison keysha256:b26f8a3d2606999e…

sourceLiger-Kernel benchmarks

external idfused_neighborhood_attention/forward/torch/nvidia-h100-80gb-hbm3/batch4-dilation1-heads12-hidden768-kernel_size7-seq256-fp32/2025-05-27-19-25-47

sha256:9311abdac89a30cab85260e…

Correctness

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

Workload

seq256

batch4

heads12

hidden768

dilation1

kernel_size7

k_biasfp32 [768]

q_biasfp32 [768]

v_biasfp32 [768]

k_weightfp32 [768, 768]

out_biasfp32 [768]

q_weightfp32 [768, 768]

v_weightfp32 [768, 768]

out_weightfp32 [768, 768]

hidden_statesfp32 [4, 256, 768]

definition comparatornot_asserted

Measurements

latency · median20.9 ms

latency · p2020.9 ms

latency · p8021.3 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": 20872080
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2025-05-27T19:25:47.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 20860186,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 21320633,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 20.860185623168945,
          "ms_50": 20.872079849243164,
          "ms_80": 21.320632934570312
        },
        "benchmark": "fused_neighborhood_attention/forward",
        "externalId": "fused_neighborhood_attention/forward/torch/nvidia-h100-80gb-hbm3/batch4-dilation1-heads12-hidden768-kernel_size7-seq256-fp32/2025-05-27-19-25-47"
      },
      "protocolDigest": "sha256:a79eaf4df66c97c6adc1fb2d09df78682155d1ea86652ae0b70199875bde7747",
      "workloadDigest": "sha256:4ca9a8f60b9cbc20eb5b9f6b7a227acace8b91f7ee13f6738a8941f3d14c3ce2",
      "environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
      "implementationDigest": "sha256:0c3e4e2a4849358a1d5183daed7f408a3422f6cc411cc6bd7d4ebeead8e57e00"
    },
    "metadata": {
      "name": "liger-fused-neighborhood-attention-forward-torch-nvidia-h100-80gb-hbm3-batch4-dilation1-heads12-hidd",
      "title": "Fused neighborhood attention · torch · forward",
      "labels": {
        "liger_version": "0.5.10"
      }
    },
    "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 module forward pass, 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-forward",
      "title": "Liger benchmark harness · forward 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 neighborhood attention →JSON