Run 01a0202b · KernelIndex

Primary measurement

7.74msmedian

Rank 1 in its comparison group · source-native comparison · observed 2024-12-03

ms 20 7.729177474975586 · ms 50 7.735536098480225 · ms 80 7.798131465911865

Identity

implementationLiger

projectLiger-Kernel

revisionunknown

workloadvocab = 128256 · hidden = 4096 · tokens = 1024 · bf16

comparison keysha256:a2949f3ae8a327df…

sourceLiger-Kernel benchmarks

external iddistill_jsd_loss/forward/liger/nvidia-h100-80gb-hbm3/hidden4096-tokens1024-vocab128256-bf16/2024-12-03-07-58-46

sha256:6d3af5006fbc755c35d5f75…

Correctness

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

Workload

vocab128256

hidden4096

tokens1024

targetint64 [1024]

student_inputbf16 [1024, 2048]

teacher_inputbf16 [1024, 4096]

student_weightbf16 [128256, 2048]

teacher_weightbf16 [128256, 4096]

definition comparatornot_asserted

Measurements

latency · median7.74 ms

latency · p207.73 ms

latency · p807.80 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": 7735536
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2024-12-03T07:58:46.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 7729177,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 7798131,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 7.729177474975586,
          "ms_50": 7.735536098480225,
          "ms_80": 7.798131465911865
        },
        "benchmark": "distill_jsd_loss/forward",
        "externalId": "distill_jsd_loss/forward/liger/nvidia-h100-80gb-hbm3/hidden4096-tokens1024-vocab128256-bf16/2024-12-03-07-58-46"
      },
      "protocolDigest": "sha256:a79eaf4df66c97c6adc1fb2d09df78682155d1ea86652ae0b70199875bde7747",
      "workloadDigest": "sha256:f42c9ab5197b3f4089634fb6948990c4ab08c98ef993934d50fe2e17fcf9ebf3",
      "environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
      "implementationDigest": "sha256:d837dbdac56dc2e7711f8e547e5951891cab10ff477f4f1fac1a7982fdb1dde8"
    },
    "metadata": {
      "name": "liger-distill-jsd-loss-forward-liger-nvidia-h100-80gb-hbm3-hidden4096-tokens1024-vocab128256-bf16-20",
      "title": "Chunked distillation JSD loss · liger · forward",
      "labels": {
        "liger_version": "0.4.2"
      }
    },
    "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 Chunked distillation JSD loss →JSON