Skip to content
KernelIndex
Search⌘K

Rapid Impala

GQA paged decode h32 kv8 d128 ps1 · suite of 48 cases · mean latency · run 01a03529-99c4…
●passedReported evidence
Primary measurement
126.1µsmean

Rank 16 in its comparison group · source-native comparison · observed 2026-07-06

Compare with #1 →

SOL 0.4832 · 1.16× reference · 27/48 cases faster

Identity
implementationRapid Impala
projectRapid Impala
revisionsubmission-17586
workloadsuite of 48 cases · mean latency
comparison keysha256:333c3cf72b3900a7…
sourceNVIDIA SOL-ExecBench
external idsubmission/17586
sha256:3d72856f8a576f0301e8127…

Correctness

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

Workload

definition comparatorsol_execbench_eval
descriptionCorrectness must pass on every case in the SOL evaluation stack.

Measurements

latency · mean126.1 µs

Protocol

harnessSOL-ExecBench evaluation stack v1.0
timerharness_reported
primaryStatisticmean
correctnessReferencedefinition reference implementation
comparabilityFamilysol_execbench_leaderboard

Environment

gpuNVIDIA B200 (sm_100)

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": {
        "latencyNs": {
          "mean": 126073
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2026-07-06T08:23:36.141Z",
      "sourceNative": {
        "source": "sol-execbench",
        "metrics": {
          "sol_score": 0.483172,
          "latency_ms": 0.126073,
          "avg_speedup": 1.155,
          "fast_1_count": 27,
          "fast_1_total": 48
        },
        "benchmark": "013_gqa_paged_decode_h32_kv8_d128_ps1",
        "externalId": "17586"
      },
      "protocolDigest": "sha256:2964c5c4ed9cb0dcdc0f74674035bc71bf5e66620939e38441060f76434ed3e5",
      "workloadDigest": "sha256:f790bd774067e85b61effaf0a712ec16ee92e0ddd0edba5180529bf0a89ef8c1",
      "environmentDigest": "sha256:5296c820e166474e590abb1beaf7728cdfe0a4bde9bf55f8966a40f4f5401621",
      "implementationDigest": "sha256:3f236e2c95700a3dc29b06c00619d831daa0ff91560f86f39e644da3e5ce7831"
    },
    "metadata": {
      "name": "sol-submission-17586",
      "title": "Rapid Impala · 013_gqa_paged_decode_h32_kv8_d128_ps1 · B200",
      "labels": {
        "worker": "B200-ext-prod-2954"
      }
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "protocol": {
    "kind": "BenchmarkProtocol",
    "spec": {
      "harness": {
        "name": "SOL-ExecBench evaluation stack",
        "version": "v1.0"
      },
      "correctness": {
        "reference": "definition reference implementation",
        "comparator": "sol_execbench_eval"
      },
      "measurement": {
        "timer": "harness_reported",
        "primaryStatistic": "mean"
      },
      "comparability": {
        "notes": "Suite-mean latency and SOL-Score as published by the SOL-ExecBench leaderboard. Comparable only within one definition, GPU type, and evaluation stack version.",
        "family": "sol_execbench_leaderboard"
      }
    },
    "metadata": {
      "name": "sol-execbench-leaderboard-v1-0",
      "title": "SOL-ExecBench evaluation stack v1.0"
    },
    "apiVersion": "kernelindex.dev/v1alpha1"
  },
  "environment": {
    "kind": "ExecutionEnvironment",
    "spec": {
      "hardware": {
        "vendor": "nvidia",
        "product": "NVIDIA B200",
        "architecture": "sm_100"
      },
      "software": {
        "libraries": {
          "evaluation_stack": "v1.0"
        }
      }
    },
    "metadata": {
      "name": "sol-execbench-b200-v1-0",
      "title": "SOL-ExecBench NVIDIA B200, stack v1.0"
    },
    "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 2026-08-24 · NVIDIA SOL-ExecBenchAll results for GQA paged decode h32 kv8 d128 ps1 →JSON