Run 01a02032 · KernelIndex

Primary measurement

3.72msmean

Rank 4 in its comparison group · source-native comparison · observed 2025-10-16

Compare with #1 →

3.77× reference

Identity

implementationgpt-5_triton_a41cd4

projectgpt-5-2025-08-07

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadnum_pages = 989669 · batch_size = 16 · len_indptr = 17 · num_kv_indices = 2757 · bf16

comparison keysha256:cb9c281ee2605ed4…

sourceFlashInfer-Bench

external idtrace/gpt-5_triton_a41cd4/5e86857b-bb94-4f2c-b9e6-6000c253dc10

sha256:30d4b0bd4f5883e20053cb2…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.0078125

max relative error0.41109445691108704

resultpassed

Workload

num_pages989669

batch_size16

len_indptr17

num_kv_indices2757

q_pebf16 [16, 16, 64]

q_nopebf16 [16, 16, 512]

ckv_cachebf16 [989669, 1, 512]

kpe_cachebf16 [989669, 1, 64]

kv_indptrint32 [17]

kv_indicesint32 [2757]

definition comparatorsol_execbench_eval

Measurements

latency · mean3.72 ms

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": 3722459
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2025-10-16T11:05:30.701Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.0078125,
        "maximumRelativeError": 0.41109445691108704
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 3.722459059495192,
          "speedup_factor": 3.772858957604076,
          "reference_latency_ms": 14.04431300693088
        },
        "benchmark": "mla_paged_decode_h16_ckv512_kpe64_ps1"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:22c02d5fd2b8566217785929bf44e60df426601978916d0727514ca3dde6febe",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:c4d597804add5bfbd3c24de72e5a7c66c9f333e568fe8cdfcc9b7346fcc4cbd0"
    },
    "metadata": {
      "name": "sol-trace-mla-paged-decode-h16-ckv512-kpe64-ps1-5e86857b",
      "title": "gpt-5_triton_a41cd4 · mla_paged_decode_h16_ckv512_kpe64_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 MLA paged decode h16 ckv512 kpe64 ps1 →JSON

gpt-5 / tritona41cd4