Run 01a00b1d · KernelIndex

Primary measurement

376.6µsmean

Rank 1 in its comparison group · source-native comparison · observed 2025-10-21

154.06× reference

Identity

implementationflashinfer_wrapper_71bd33

projectFlashInfer-Bench baselines

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadtotal_q = 15344 · num_pages = 556622 · len_indptr = 28 · num_kv_indices = 48 · bf16

comparison keysha256:58163e183fdd8635…

sourceFlashInfer-Bench

external idtrace/flashinfer_wrapper_71bd33/6febcf2f-ced2-4f73-b780-18a83c327879

sha256:38b9dba10bba8f7a9f451e6…

Correctness

comparatorflashinfer_bench_eval

max relative error0

resultpassed

Workload

total_q15344

num_pages556622

len_indptr28

num_kv_indices48

qbf16 [15344, 32, 128]

k_cachebf16 [556622, 1, 4, 128]

v_cachebf16 [556622, 1, 4, 128]

kv_indptrint32 [28]

qo_indptrint32 [28]

kv_indicesint32 [48]

definition comparatorsol_execbench_eval

Measurements

latency · mean376.6 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

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": 376609
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2025-10-21T13:35:41.801Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumRelativeError": 0
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 0.376609131962367,
          "speedup_factor": 154.0637461662201,
          "reference_latency_ms": 58.0218137105306
        },
        "benchmark": "gqa_paged_prefill_causal_h32_kv4_d128_ps1"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:295a39b0233afc8fa5c80dde9b7f302f620386a8ebe1c58a56f2c0b7cb3c6726",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:23fde8c921c90d6156a877704f9e7329fdc0550b874f804f2b2992c6fe4a2bd5"
    },
    "metadata": {
      "name": "sol-trace-gqa-paged-prefill-causal-h32-kv4-d128-ps1-6febcf2f",
      "title": "flashinfer_wrapper_71bd33 · gqa_paged_prefill_causal_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 2026-08-16 · FlashInfer-Bench · Apache-2.0All results for GQA paged prefill causal h32 kv4 d128 ps1 →JSON

flashinfer / wrapper71bd33