Liger

{
  "run": {
    "kind": "BenchmarkRun",
    "spec": {
      "status": "passed",
      "timing": {
        "latencyNs": {
          "median": 53248
        },
        "primaryStatistic": "median"
      },
      "observedAt": "2025-04-30T16:11:10.000Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 45056,
          "metric": "latency",
          "statistic": "p20"
        },
        {
          "unit": "ns",
          "value": 61722,
          "metric": "latency",
          "statistic": "p80"
        }
      ],
      "sourceNative": {
        "source": "liger-kernel-bench",
        "metrics": {
          "ms_20": 0.04505600035190582,
          "ms_50": 0.053247999399900436,
          "ms_80": 0.06172160431742668
        },
        "benchmark": "softmax/full",
        "externalId": "softmax/full/liger/nvidia-geforce-rtx-3090/hidden128-rows2048-fp32/2025-04-30-16-11-10"
      },
      "protocolDigest": "sha256:5b02ee9a2140fbc6fcc1c89800f6c177ba22718c8897cb532bf0ba91867c1e58",
      "workloadDigest": "sha256:a95cdf9fdf1be9a0a58acbe6d2e7381f4d43858343d1f6e08f6072b783823577",
      "environmentDigest": "sha256:b12fdfab6a337672499b10212a1e986e49e4383777bd7cbb58321b09af4bf6a3",
      "implementationDigest": "sha256:e13f772e1dd8b15a7c1f8d2ab7fe54394500d7ace42648fc352d9ca36c0f363a"
    },
    "metadata": {
      "name": "liger-softmax-full-liger-nvidia-geforce-rtx-3090-hidden128-rows2048-fp32-2025-04-30-16-11-10",
      "title": "Softmax · liger · full",
      "labels": {
        "liger_version": "0.5.8"
      }
    },
    "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 GeForce RTX 3090",
        "architecture": "sm_86"
      },
      "software": {}
    },
    "metadata": {
      "name": "liger-bench-nvidia-geforce-rtx-3090",
      "title": "Liger benchmark host · GeForce RTX 3090"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}