Transformers
Rotary position embedding · seq = 2048 · hidden = 512 · q_heads = 32 · kv_heads = 8 · bf16 · run 01a016ae-0dfb…
●passedReported evidence
Not re-observed recently.
Primary measurement
79.7µsmedian
Rank 2 in its comparison group · source-native comparison · observed 2024-09-03
ms 20 0.07923199981451035 · ms 50 0.07968000322580338 · ms 80 0.10408961027860641
Identity
implementationTransformers
projectHugging Face Transformers
revisionunknown
workloadseq = 2048 · hidden = 512 · q_heads = 32 · kv_heads = 8 · bf16
comparison keysha256:ff31972e560bf3b4…
sourceLiger-Kernel benchmarks
external idrope/forward/huggingface/nvidia-a100-sxm4-80gb/hidden512-kv_heads8-q_heads32-seq2048-bf16/2024-09-03-15-39-56
sha256:3c3bdc3c7537de317f4d001…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
seq2048
hidden512
q_heads32
kv_heads8
kbf16 [1, 8, 2048, 16]
qbf16 [1, 32, 2048, 16]
definition comparatornot_asserted
Measurements
latency · median79.7 µs
latency · p2079.2 µs
latency · p80104.1 µs
Protocol
harnessLiger-Kernel benchmark scripts
timercuda_events
primaryStatisticmedian
comparabilityFamilyliger_kernel_bench
Environment
gpuNVIDIA A100 (sm_80)
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": {
"median": 79680
},
"primaryStatistic": "median"
},
"observedAt": "2024-09-03T15:39:56.000Z",
"measurements": [
{
"unit": "ns",
"value": 79232,
"metric": "latency",
"statistic": "p20"
},
{
"unit": "ns",
"value": 104090,
"metric": "latency",
"statistic": "p80"
}
],
"sourceNative": {
"source": "liger-kernel-bench",
"metrics": {
"ms_20": 0.07923199981451035,
"ms_50": 0.07968000322580338,
"ms_80": 0.10408961027860641
},
"benchmark": "rope/forward",
"externalId": "rope/forward/huggingface/nvidia-a100-sxm4-80gb/hidden512-kv_heads8-q_heads32-seq2048-bf16/2024-09-03-15-39-56"
},
"protocolDigest": "sha256:a79eaf4df66c97c6adc1fb2d09df78682155d1ea86652ae0b70199875bde7747",
"workloadDigest": "sha256:1415fc0b1b53aa8f16a41dfe92a66a1584b2d0b9523b0058543ea3a8ac7b5314",
"environmentDigest": "sha256:90c9e2764222f592477cfccf3373519add5a1b881563956ad058c4f99e41b6b9",
"implementationDigest": "sha256:459baf6556e5536285c26b26d11eda4980b44d362142461d8e7e5b4f3900ce23"
},
"metadata": {
"name": "liger-rope-forward-huggingface-nvidia-a100-sxm4-80gb-hidden512-kv-heads8-q-heads32-seq2048-bf16-2024",
"title": "Rotary position embedding · huggingface · forward",
"labels": {
"liger_version": "0.2.1"
}
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"protocol": {
"kind": "BenchmarkProtocol",
"spec": {
"harness": {
"name": "Liger-Kernel benchmark scripts",
"repository": "https://github.com/linkedin/Liger-Kernel"
},
"measurement": {
"timer": "cuda_events",
"primaryStatistic": "median"
},
"comparability": {
"notes": "Timed: the module forward pass, via triton.testing.do_bench with quantiles 0.5/0.2/0.8 (median with p20/p80 spread). The upstream CSV records no CUDA, driver, or torch version; the Liger release rides each run's labels. Comparable only within one kernel, workload, timed pass, and GPU.",
"family": "liger_kernel_bench"
}
},
"metadata": {
"name": "liger-bench-do-bench-forward",
"title": "Liger benchmark harness · forward pass"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"environment": {
"kind": "ExecutionEnvironment",
"spec": {
"hardware": {
"vendor": "nvidia",
"product": "NVIDIA A100",
"formFactor": "SXM4-80GB",
"architecture": "sm_80"
},
"software": {}
},
"metadata": {
"name": "liger-bench-nvidia-a100-sxm4-80gb",
"title": "Liger benchmark host · A100-SXM4-80GB"
},
"apiVersion": "kernelindex.dev/v1alpha1"
}
}Cite this record (permalink, digest, access date)
Report an issue with this run
Published 2026-08-18 · Liger-Kernel benchmarks · BSD-2-ClauseAll results for Rotary position embedding →JSON