Run 01a02032 · KernelIndex

Primary measurement

4.99msmean

Rank 1 in its comparison group · source-native comparison · observed 2025-10-16

2.63× reference

Identity

implementationgpt-5-2025-08-07_cuda_a2d8ca

projectgpt-5-2025-08-07

revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c

workloadseq_len = 16 · bf16/fp32/fp8_e4m3

comparison keysha256:256a1572e32e3658…

sourceFlashInfer-Bench

external idtrace/gpt-5-2025-08-07_cuda_a2d8ca/a7c2bcfd-a2f4-479e-8d32-200115df89cf

sha256:b9e0348d6c4212c7cbbab5f…

Correctness

comparatorflashinfer_bench_eval

max absolute error0.0000019073486328125

max relative error0.006711172871291637

resultpassed

Workload

seq_len16

routing_biasbf16 [256]

gemm1_weightsfp8_e4m3 [32, 4096, 7168]

gemm2_weightsfp8_e4m3 [32, 7168, 2048]

hidden_statesfp8_e4m3 [16, 7168]

routing_logitsfp32 [16, 256]

gemm1_weights_scalefp32 [32, 32, 56]

gemm2_weights_scalefp32 [32, 56, 16]

hidden_states_scalefp32 [56, 16]

definition comparatorsol_execbench_eval

Measurements

latency · mean4.99 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": 4993890
        },
        "primaryStatistic": "mean"
      },
      "observedAt": "2025-10-16T12:03:11.456Z",
      "correctness": {
        "comparator": "sol_execbench_eval",
        "maximumAbsoluteError": 0.0000019073486328125,
        "maximumRelativeError": 0.006711172871291637
      },
      "sourceNative": {
        "source": "flashinfer-bench",
        "metrics": {
          "latency_ms": 4.993890320813215,
          "speedup_factor": 2.6327929181602068,
          "reference_latency_ms": 13.147879070705836
        },
        "benchmark": "moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048"
      },
      "protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
      "workloadDigest": "sha256:0e02b21e52e6f91289b51c444f7ff5741f2be70580abec40281d6956641f4de9",
      "environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
      "implementationDigest": "sha256:17a59015a553f958aa6ed224d10af90a5700205fe68519dbeaf081a7ebffb832"
    },
    "metadata": {
      "name": "sol-trace-moe-fp8-block-scale-ds-routing-topk8-ng8-kg4-e32-h7168-i2048-a7c2bcfd",
      "title": "gpt-5-2025-08-07_cuda_a2d8ca · moe_fp8_block_scale_ds_routing_topk8_ng8_kg4_e32_h7168_i2048"
    },
    "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 MoE FP8 block scale ds routing topk8 ng8 kg4 e32 h7168 i2048 →JSON

gpt-5-2025-08-07 / cudaa2d8ca