Skip to content
KernelIndex
Search⌘K

submission 30355

AMD FP8 blockwise GEMM · k = 7168 · m = 6144 · n = 4608 · seed = 65436 · fp32/fp8_e4m3fnuz · run 01a006cc-a9f1…
●passedReported evidence

Not re-observed recently.

Primary measurement
652.9µsmean of 100

Rank 15 in its comparison group · source-native comparison · observed 2025-06-01

Compare with #1 →

best ns 604308 · mean ns 652873.28 · leaderboard score s 0.0001335604527094927

Identity
implementationsubmission 30355
projectKernelBot user 415
revisionsubmission-30355
workloadk = 7168 · m = 6144 · n = 4608 · seed = 65436 · fp32/fp8_e4m3fnuz
comparison keysha256:2a59e128cd52b115…
sourceGPU MODE KernelBot
external idsubmission/30355/benchmark/14
sha256:61ae949d2e977eaaaec0a66…

Correctness

Marked passed by the source; the correctness policy was not published.

Workload

k7168
m6144
n4608
seed65436
afp8_e4m3fnuz [6144, 7168]
bfp8_e4m3fnuz [4608, 7168]
a_scalefp32 [6144, 56]
b_scalefp32 [36, 56]
definition comparatorkernelbot_eval

Measurements

latency · max728.0 µs · n=100
latency · mean652.9 µs · n=100
latency · min604.3 µs · n=100
latency · std29.9 µs · n=100

Protocol

harnessreference-kernels eval.py
timerwall_clock
primaryStatisticmean
correctnessReferenceproblem reference implementation
comparabilityFamilygpumode_kernelbot

Environment

gpuAMD Instinct MI300X (gfx942)

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
I

Published under your account name. Community attestations never change the evidence level; a maintainer can hide abuse.

Canonical manifest

Show manifest
{
  "run": {
    "kind": "BenchmarkRun",
    "spec": {
      "status": "passed",
      "timing": {
        "samples": 100,
        "latencyNs": {
          "mean": 652873,
          "maximum": 728000,
          "minimum": 604308
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2025-06-01T23:41:12.770Z",
      "measurements": [
        {
          "unit": "ns",
          "value": 29873.931448733627,
          "metric": "latency",
          "statistic": "std",
          "sampleCount": 100
        }
      ],
      "sourceNative": {
        "source": "gpumode-kernelbot",
        "metrics": {
          "best_ns": 604308,
          "mean_ns": 652873.28,
          "leaderboard_score_s": 0.0001335604527094927
        },
        "benchmark": "amd-fp8-mm/14",
        "externalId": "submission/30355"
      },
      "protocolDigest": "sha256:7c66da4b5866502eb10444a7ba013d0f6bdf1620c55847945008cad2fbf98aec",
      "workloadDigest": "sha256:7b85796aca4f96a8793730a378dcce0bb1a9e7415b11867871ffab0b1643063c",
      "environmentDigest": "sha256:f722892449ed976f3e21b12df8b92b34573fc232883fec6a2f3da399a00abdd5",
      "implementationDigest": "sha256:78f89793ec42277e6e7f14141d02bd40e1f0ad8bed767b08b80b419754176510"
    },
    "metadata": {
      "name": "kernelbot-30355-k7168-m6144-n4608-seed65436",
      "title": "amd-fp8-mm · submission 30355 · case 14",
      "labels": {
        "torch": "2.8.0.dev20250530+rocm6.3",
        "platform": "Linux-5.15.0-136-generic-x86_64-with-glibc2.35"
      }
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "protocol": {
    "kind": "BenchmarkProtocol",
    "spec": {
      "harness": {
        "name": "reference-kernels eval.py",
        "repository": "https://github.com/gpu-mode/reference-kernels"
      },
      "correctness": {
        "reference": "problem reference implementation",
        "comparator": "kernelbot_eval"
      },
      "measurement": {
        "timer": "wall_clock",
        "primaryStatistic": "mean"
      },
      "comparability": {
        "notes": "Per-case wall-clock statistics as published in the KernelBot dataset. Comparable only within one leaderboard problem and GPU type; runner software versions ride each run's labels.",
        "family": "gpumode_kernelbot"
      }
    },
    "metadata": {
      "name": "gpumode-kernelbot-eval",
      "title": "GPU MODE KernelBot evaluation harness"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "environment": {
    "kind": "ExecutionEnvironment",
    "spec": {
      "hardware": {
        "vendor": "amd",
        "product": "AMD Instinct MI300X",
        "architecture": "gfx942"
      },
      "software": {}
    },
    "metadata": {
      "name": "gpumode-kernelbot-amd-instinct-mi300x",
      "title": "KernelBot runner · AMD Instinct MI300X"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  }
}
Cite this record (permalink, digest, access date)
Report an issue with this run

A maintainer reviews every report. If accepted, the record is retracted or superseded; its history stays visible.

Published 2026-08-15 · GPU MODE KernelBot · June 9 Researcher Reciprocity License v1.0All results for AMD FP8 blockwise GEMM →JSON