gpt-o3 / cudaa7bbcf
Fused add RMSNorm h4096 · batch_size = 34 · bf16 · run 01a02032-5e4c…
●passedReproduction-ready evidence
Not re-observed recently.
Primary measurement
17.8µsmean
Rank 5 in its comparison group · source-native comparison · observed 2025-10-16
67.46× reference
Identity
implementationgpt-o3_cuda_a7bbcf
projectgpt-o3
revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c
workloadbatch_size = 34 · bf16
comparison keysha256:fe7a9eb74af889f1…
sourceFlashInfer-Bench
external idtrace/gpt-o3_cuda_a7bbcf/04e1c8da-5407-4d96-876c-a27259fe7ecf
sha256:4d1c6401149d947435e68d2…
Correctness
comparatorflashinfer_bench_eval
max absolute error0
max relative error0
resultpassed
Workload
batch_size34
weightbf16 [4096]
residualbf16 [34, 4096]
hidden_statesbf16 [34, 4096]
definition comparatorsol_execbench_eval
Measurements
latency · mean17.8 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
logssha256:5c2bfaa09540be51… · 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 manifestHide manifest
{
"run": {
"kind": "BenchmarkRun",
"spec": {
"status": "passed",
"timing": {
"latencyNs": {
"mean": 17805
},
"primaryStatistic": "mean"
},
"evidence": {
"logs": {
"uri": "kernelindex:artifact/sha256:5c2bfaa09540be51d7764c6ece42106b2f8cdda86cd622ffd7a11a6c86776e4a",
"digest": "sha256:5c2bfaa09540be51d7764c6ece42106b2f8cdda86cd622ffd7a11a6c86776e4a"
}
},
"observedAt": "2025-10-16T12:07:45.283Z",
"correctness": {
"comparator": "sol_execbench_eval",
"maximumAbsoluteError": 0,
"maximumRelativeError": 0
},
"sourceNative": {
"source": "flashinfer-bench",
"metrics": {
"latency_ms": 0.01780510653298123,
"speedup_factor": 67.46075746186506,
"reference_latency_ms": 1.2011459734041157
},
"benchmark": "fused_add_rmsnorm_h4096"
},
"protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
"workloadDigest": "sha256:30c05d7aabf935d0cdc8d6917badf3c05d6aac124a811b4b75309799728413ff",
"environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
"implementationDigest": "sha256:8c12465bcf5c44347374419198e781a530c519953fc712d53eb48aff302248b8"
},
"metadata": {
"name": "sol-trace-fused-add-rmsnorm-h4096-04e1c8da",
"title": "gpt-o3_cuda_a7bbcf · fused_add_rmsnorm_h4096"
},
"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 Fused add RMSNorm h4096 →JSON