Run 01a00b1d · KernelIndex

Primary measurement

146.3µsmean

Rank 2 in its comparison group · source-native comparison · observed 2026-03-28

Compare with #1 →

23.18× reference

Identity

implementationflashinfer_wrapper_94b73a

projectFlashInfer-Bench baselines

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadnum_pages = 327 · batch_size = 1 · len_indptr = 2 · num_kv_indices = 296 · bf16

comparison keysha256:5cc2cc328f8e0498…

sourceFlashInfer-Bench

external idtrace/flashinfer_wrapper_94b73a/b018c365-63a4-4a45-a744-1363315d413a

sha256:2771b53718299ffc998fbbc…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.001953125

max relative error4.185848236083984

resultpassed

Workload

num_pages327

batch_size1

len_indptr2

num_kv_indices296

qbf16 [1, 20, 128]

k_cachebf16 [327, 1, 4, 128]

v_cachebf16 [327, 1, 4, 128]

kv_indptrint32 [2]

kv_indicesint32 [296]

definition comparatorsol_execbench_eval

Measurements

latency · mean146.3 µs

Protocol

harnessflashinfer-bench

timerharness_reported

primaryStatisticmean

correctnessReferencedefinition reference implementation

comparabilityFamilyflashinfer_bench_trace

Environment

gpuNVIDIA B200 (unknown)

cudaToolkit12.8

frameworkpytorch 2.9.1+cu128

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": {
          "mean": 146272
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2026-03-28T04:29:17.306Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.001953125,
        "maximumRelativeError": 4.185848236083984
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 0.146272,
          "speedup_factor": 23.179282432728066,
          "reference_latency_ms": 3.39048
        },
        "benchmark": "gqa_paged_decode_h20_kv4_d128_ps1"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:afd1416cfe0f418ed61ffa69483021b8df1d79597c3218fec859d10cc4029209",
      "environmentDigest": "sha256:af7cb68a0bb0eb9671092a620668c2600f0431301fc46a71807ea75fb7de8bce",
      "implementationDigest": "sha256:7315312046df01edd99c862f46bcc06a77f30e59a7b8416fd23df5d760a0e9da"
    },
    "metadata": {
      "name": "sol-trace-gqa-paged-decode-h20-kv4-d128-ps1-b018c365",
      "title": "flashinfer_wrapper_94b73a · gqa_paged_decode_h20_kv4_d128_ps1"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "protocol": {
    "kind": "BenchmarkProtocol",
    "spec": {
      "harness": {
        "name": "flashinfer-bench",
        "repository": "https://github.com/flashinfer-ai/flashinfer-bench"
      },
      "correctness": {
        "reference": "definition reference implementation",
        "comparator": "flashinfer_bench_eval"
      },
      "measurement": {
        "timer": "harness_reported",
        "primaryStatistic": "mean"
      },
      "comparability": {
        "family": "flashinfer_bench_trace"
      }
    },
    "metadata": {
      "name": "flashinfer-bench-trace",
      "title": "FlashInfer-Bench trace harness"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "environment": {
    "kind": "ExecutionEnvironment",
    "spec": {
      "hardware": {
        "vendor": "nvidia",
        "product": "NVIDIA B200",
        "architecture": "unknown"
      },
      "software": {
        "framework": {
          "name": "pytorch",
          "version": "2.9.1+cu128"
        },
        "libraries": {
          "triton": "3.5.1"
        },
        "cudaToolkit": "12.8"
      }
    },
    "metadata": {
      "name": "sol-trace-nvidia-b200-2-9-1-cu128-3-5-1-12-8",
      "title": "NVIDIA B200 (trace-reported software)"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}

Cite this record (permalink, digest, access date)

Report an issue with this run

Published 2026-08-16 · FlashInfer-Bench · Apache-2.0All results for GQA paged decode h20 kv4 d128 ps1 →JSON

flashinfer / wrapper94b73a