Run 01a02032 · KernelIndex

Primary measurement

319.4µsmean

Rank 4 in its comparison group · source-native comparison · observed 2025-10-20

Compare with #1 →

1.38× reference

Identity

implementationgpt-o3_triton_a47205

projectgpt-o3

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadm = 70 · fp16

comparison keysha256:68ea3e99d840d533…

sourceFlashInfer-Bench

external idtrace/gpt-o3_triton_a47205/c1e3dcee-7b0f-4e63-b99e-1666ced4cbb2

sha256:34c7d49aa41ee88ab001bb7…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.25

max relative error7.124533176422119

resultpassed

Workload

m70

afp16 [70, 14336]

bfp16 [4096, 14336]

definition comparatorsol_execbench_eval

Measurements

latency · mean319.4 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": 319371
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2025-10-20T03:27:06.078Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.25,
        "maximumRelativeError": 7.124533176422119
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 0.3193714251147678,
          "speedup_factor": 1.3820595446594737,
          "reference_latency_ms": 0.44139032637136316
        },
        "benchmark": "gemm_n4096_k14336"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:d34dde80dd0e71e04c7e909f917dd4c7db833dac0ba36760ca50ab3496325683",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:2347747be2b550c4de9eba44a19d87ae08554e8d4548b1c73a15f3e2cf795b71"
    },
    "metadata": {
      "name": "sol-trace-gemm-n4096-k14336-c1e3dcee",
      "title": "gpt-o3_triton_a47205 · gemm_n4096_k14336"
    },
    "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 GEMM n4096 k14336 →JSON