Run 01a02032 · KernelIndex

Primary measurement

175.2µsmean

Rank 6 in its comparison group · source-native comparison · observed 2025-10-16

Compare with #1 →

185.30× reference

Identity

implementationgpt-5_cuda_00b2dd

projectgpt-5-2025-08-07

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadnum_pages = 191 · batch_size = 1 · len_indptr = 2 · num_kv_indices = 141 · bf16

comparison keysha256:bb088a84e0dedfda…

sourceFlashInfer-Bench

external idtrace/gpt-5_cuda_00b2dd/4b9dd400-879c-4797-9187-3cd546d6dd1b

sha256:a27167de2122edaf347ee48…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.000244140625

max relative error0.02660597302019596

resultpassed

Workload

num_pages191

batch_size1

len_indptr2

num_kv_indices141

qbf16 [1, 32, 128]

k_cachebf16 [191, 1, 4, 128]

v_cachebf16 [191, 1, 4, 128]

kv_indptrint32 [2]

kv_indicesint32 [141]

definition comparatorsol_execbench_eval

Measurements

latency · mean175.2 µs

Protocol

harnessflashinfer-bench

timerharness_reported

primaryStatisticmean

correctnessReferencedefinition reference implementation

comparabilityFamilyflashinfer_bench_trace

Environment

gpuNVIDIA B200 (unknown)

cudaToolkit12.8

frameworkpytorch 2.8.0+cu128

Artifacts

logssha256:2ef5ad2323f2e321… · text/plain · unavailable

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": 175201
        },
        "primaryStatistic": "mean"
      },
      "evidence": {
        "logs": {
          "uri": "kernelindex:artifact/sha256:2ef5ad2323f2e321bf33fd4dc3577c1696542bc494283492746b360d4638963c",
          "digest": "sha256:2ef5ad2323f2e321bf33fd4dc3577c1696542bc494283492746b360d4638963c"
        }
      },
      "observedAt": "2025-10-16T01:25:48.630Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.000244140625,
        "maximumRelativeError": 0.02660597302019596
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 0.17520112264628848,
          "speedup_factor": 185.2987193810628,
          "reference_latency_ms": 32.46454366048177
        },
        "benchmark": "gqa_paged_decode_h32_kv4_d128_ps1"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:3d7681117468bbad4394a515924e5f94c74d91639ece60ff07c6309c8204bde1",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:3b7023b82a8abc5ad4f3b833025c2f32a6e068b4c14d60cb171d681a82b194ed"
    },
    "metadata": {
      "name": "sol-trace-gqa-paged-decode-h32-kv4-d128-ps1-4b9dd400",
      "title": "gpt-5_cuda_00b2dd · gqa_paged_decode_h32_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": "2.8.0+cu128"
        },
        "libraries": {
          "triton": "3.4.0"
        },
        "cudaToolkit": "12.8"
      }
    },
    "metadata": {
      "name": "sol-trace-nvidia-b200-2-8-0-cu128-3-4-0-12-8",
      "title": "NVIDIA B200 (trace-reported software)"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}

Cite this record (permalink, digest, access date)

Report an issue with this run

Published 2026-08-20 · FlashInfer-Bench · Apache-2.0All results for GQA paged decode h32 kv4 d128 ps1 →JSON

gpt-5 / cuda00b2dd