Run 01a0202b · KernelIndex

Primary measurement

780.1µsmedian

Rank 2 in its comparison group · source-native comparison · observed 2025-08-07

Compare with #1 →

ms 20 0.7790719866752625 · ms 50 0.7800959944725037 · ms 80 0.7810239791870117

Identity

implementationTransformers

projectHugging Face Transformers

revisionunknown

workloadseq = 4096 · hidden = 8192 · q_heads = 32 · kv_heads = 8 · bf16

comparison keysha256:276b7468409a9626…

sourceLiger-Kernel benchmarks

external idllama4_rope/full/huggingface/nvidia-h100-80gb-hbm3/hidden8192-kv_heads8-q_heads32-seq4096-bf16/2025-08-07-21-42-21

sha256:d98e979a68b1308e17e9a5f…

Correctness

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

Workload

seq4096

hidden8192

q_heads32

kv_heads8

kbf16 [1, 4096, 8, 256]

qbf16 [1, 4096, 32, 256]

definition comparatornot_asserted

Measurements

latency · median780.1 µs

latency · p20779.1 µs

latency · p80781.0 µs

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": 780096
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2025-08-07T21:42:21.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 779072,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 781024,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 0.7790719866752625,
          "ms_50": 0.7800959944725037,
          "ms_80": 0.7810239791870117
        },
        "benchmark": "llama4_rope/full",
        "externalId": "llama4_rope/full/huggingface/nvidia-h100-80gb-hbm3/hidden8192-kv_heads8-q_heads32-seq4096-bf16/2025-08-07-21-42-21"
      },
      "protocolDigest": "sha256:5b02ee9a2140fbc6fcc1c89800f6c177ba22718c8897cb532bf0ba91867c1e58",
      "workloadDigest": "sha256:8974f9ae1a4284ed6358ddaa324eeb4c076d3c5916579ee84b38fc474a3281bf",
      "environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
      "implementationDigest": "sha256:cca024409163839a832c1d39bd301f536494b645171665d65fae5faf2717be48"
    },
    "metadata": {
      "name": "liger-llama4-rope-full-huggingface-nvidia-h100-80gb-hbm3-hidden8192-kv-heads8-q-heads32-seq4096-bf16",
      "title": "Llama 4 rotary position embedding · huggingface · full",
      "labels": {
        "liger_version": "0.6.1"
      }
    },
    "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 forward pass plus the backward 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-full",
      "title": "Liger benchmark harness · full 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 Llama 4 rotary position embedding →JSON