Run 01a02032 · KernelIndex

Primary measurement

125.5msmean

Rank 4 in its comparison group · source-native comparison · observed 2025-10-19

Compare with #1 →

0.17× reference

Identity

implementationgpt-o3_cuda_025ec1

projectgpt-o3

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadtotal_q = 12571 · total_kv = 12571 · len_indptr = 11 · bf16

comparison keysha256:b4863a05faf54deb…

sourceFlashInfer-Bench

external idtrace/gpt-o3_cuda_025ec1/7a08a0f6-7dd4-4de9-bb5b-847c51969faa

sha256:65dd812ddb481fcc8f18032…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.0078125

max relative error3.4472007751464844

resultpassed

Workload

total_q12571

total_kv12571

len_indptr11

kbf16 [12571, 8, 128]

qbf16 [12571, 32, 128]

vbf16 [12571, 8, 128]

kv_indptrint32 [11]

qo_indptrint32 [11]

definition comparatorsol_execbench_eval

Measurements

latency · mean125.5 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

logssha256:07a86bda7da2fb08… · 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": 125452993
        },
        "primaryStatistic": "mean"
      },
      "evidence": {
        "logs": {
          "uri": "kernelindex:artifact/sha256:07a86bda7da2fb08289e1ff8d4bba32c9f967ffa2ce5a6c366eb12424a4122b5",
          "digest": "sha256:07a86bda7da2fb08289e1ff8d4bba32c9f967ffa2ce5a6c366eb12424a4122b5"
        }
      },
      "observedAt": "2025-10-19T23:03:36.437Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.0078125,
        "maximumRelativeError": 3.4472007751464844
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 125.45299275716145,
          "speedup_factor": 0.16693832200786685,
          "reference_latency_ms": 20.942912101745605
        },
        "benchmark": "gqa_ragged_prefill_causal_h32_kv8_d128"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:06cc12331d1324a7050214939257d1994015345b74f4d0c04867ad4125a603cc",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:5acecad48c17870791beedfc3d3b0d9a71ca6932288f1c391313a146bb02752d"
    },
    "metadata": {
      "name": "sol-trace-gqa-ragged-prefill-causal-h32-kv8-d128-7a08a0f6",
      "title": "gpt-o3_cuda_025ec1 · gqa_ragged_prefill_causal_h32_kv8_d128"
    },
    "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 ragged prefill causal h32 kv8 d128 →JSON

gpt-o3 / cuda025ec1