gemini-2.5-pro / triton0owo26
RMSNorm h4096 · batch_size = 34 · bf16 · run 01a02032-55ac…
●passedReported evidence
Not re-observed recently.
Primary measurement
12.9µsmean
Rank 5 in its comparison group · source-native comparison · observed 2025-10-16
48.29× reference
Identity
implementationgemini-2.5-pro_triton_0owo26
projectgemini-2.5-pro
revisionda915083d4c7c5e61aa3005e3d17ae488e0fc71c
workloadbatch_size = 34 · bf16
comparison keysha256:8ff0e8feb19d3f96…
sourceFlashInfer-Bench
external idtrace/gemini-2.5-pro_triton_0owo26/f7c3b329-5d22-491c-9635-5b5906f0f4b1
sha256:73e54a0554f40b5887aa983…
Correctness
comparatorflashinfer_bench_eval
max absolute error0.001953125
max relative error0.004444444552063942
resultpassed
Workload
batch_size34
weightbf16 [4096]
hidden_statesbf16 [34, 4096]
definition comparatorsol_execbench_eval
Measurements
latency · mean12.9 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 manifestHide manifest
{
"run": {
"kind": "BenchmarkRun",
"spec": {
"status": "passed",
"timing": {
"latencyNs": {
"mean": 12935
},
"primaryStatistic": "mean"
},
"observedAt": "2025-10-16T12:16:08.611Z",
"correctness": {
"comparator": "sol_execbench_eval",
"maximumAbsoluteError": 0.001953125,
"maximumRelativeError": 0.004444444552063942
},
"sourceNative": {
"source": "flashinfer-bench",
"metrics": {
"latency_ms": 0.01293529137425714,
"speedup_factor": 48.289073712084544,
"reference_latency_ms": 0.6246332386587945
},
"benchmark": "rmsnorm_h4096"
},
"protocolDigest": "sha256:7d44d206ad086b4bece35ce31f6531ebb11f07e0d9dcfb74c75be86d3ec02095",
"workloadDigest": "sha256:d63fa1adf949e6b7db3cd30be7a69c8db1ecbc370267dbf49e6d3230fd8e8598",
"environmentDigest": "sha256:f4292a8be127b7230cd3995c1d605fb6c644e2c2f94566e1d28cc1be766ca5d5",
"implementationDigest": "sha256:c5d6dcefab0abc44d57188a1f74800b43548a83d568cb3033b4e32a30480650d"
},
"metadata": {
"name": "sol-trace-rmsnorm-h4096-f7c3b329",
"title": "gemini-2.5-pro_triton_0owo26 · 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 RMSNorm h4096 →JSON