PyTorch eager
ReLUSelfAttention · n_embd = 768 · batch_size = 16 · max_seqlen = 1024 · fp32 · run 01a03f0e-f6dd…
●passedReported evidence
Not re-observed recently.
Primary measurement
5.02ms±0.00 · mean of 100
Rank 2 in its comparison group · source-native comparison · observed 2026-03-05
max ms 5.05 · min ms 5.01 · std ms 0.00426 · mean ms 5.02
Identity
implementationPyTorch eager
projectPyTorch
revisionunknown
workloadn_embd = 768 · batch_size = 16 · max_seqlen = 1024 · fp32
comparison keysha256:5a002fb84475ea12…
sourceKernelBench baseline timings
external idH100_Modal/torch/level3/50_ReLUSelfAttention.py
sha256:66951dd33291b494c49e6be…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
n_embd768
batch_size16
max_seqlen1024
input_1fp32 [16, 1024, 768]
definition comparatornot_asserted
Measurements
latency · max5.05 ms · n=100
latency · mean5.02 ms · n=100
latency · min5.01 ms · n=100
latency · std4.26 µs
Protocol
harnessKernelBench timing scripts
timercuda_events
primaryStatisticmean
comparabilityFamilykernelbench_baseline_timing
Environment
gpuNVIDIA H100 (sm_90)
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": {
"samples": 100,
"latencyNs": {
"mean": 5020000,
"maximum": 5050000,
"minimum": 5010000,
"confidence95": [
5019165,
5020835
]
},
"primaryStatistic": "mean"
},
"observedAt": "2026-03-05T08:38:15.000Z",
"measurements": [
{
"unit": "ns",
"value": 4260,
"metric": "latency",
"statistic": "std"
}
],
"sourceNative": {
"source": "kernelbench",
"metrics": {
"max_ms": 5.05,
"min_ms": 5.01,
"std_ms": 0.00426,
"mean_ms": 5.02
},
"benchmark": "H100_Modal/baseline_time_torch.json",
"externalId": "H100_Modal/torch/level3/50_ReLUSelfAttention.py"
},
"protocolDigest": "sha256:547725b033a9911eae0c5ac362344a85aed824894ed4ab476af945b9bb491ea6",
"workloadDigest": "sha256:1862e86b68f9785737bfef55881eb54ccf50400f26c6696cbb3f94cdc608da99",
"environmentDigest": "sha256:2750ae381d6de582f1eb6b9008e0987463d19ede8af127d97d76c5ae09121393",
"implementationDigest": "sha256:0b090759fbc065eee059fe1dfa2e4cd937cd587fda5b8b39ff1ef000efb7a4d7"
},
"metadata": {
"name": "kernelbench-h100-modal-torch-l3-50-reluselfattention",
"title": "ReLUSelfAttention · PyTorch eager · Modal"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"protocol": {
"kind": "BenchmarkProtocol",
"spec": {
"harness": {
"name": "KernelBench timing scripts",
"repository": "https://github.com/ScalingIntelligence/KernelBench"
},
"measurement": {
"timer": "cuda_events",
"primaryStatistic": "mean"
},
"comparability": {
"notes": "Mean of 100 timed forward passes (CUDA events, warm-up excluded) of the reference module on fixed inputs; the 95% interval is the mean's, from the reported standard deviation. The upstream JSON records no torch or CUDA version. Comparable only within one problem and one timing host.",
"family": "kernelbench_baseline_timing"
}
},
"metadata": {
"name": "kernelbench-timing-v1",
"title": "KernelBench baseline timing"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"environment": {
"kind": "ExecutionEnvironment",
"spec": {
"hardware": {
"vendor": "nvidia",
"product": "NVIDIA H100",
"formFactor": "80GB HBM3",
"architecture": "sm_90"
},
"software": {}
},
"metadata": {
"name": "kernelbench-h100-modal",
"title": "KernelBench host · H100 80GB HBM3 (Modal)"
},
"apiVersion": "kernelindex.dev/v1alpha1"
}
}Cite this record (permalink, digest, access date)
Report an issue with this run
Published 2026-08-26 · KernelBench baseline timings · MITAll results for ReLUSelfAttention →JSON