Skip to content
KernelIndex
Search⌘K

Transformers

Hugging Face Transformers · python · Apache-2.0
holds 4 records

Use it

Source available upstream · Apache-2.0View source →

No install recipe recorded for this revision.

interfacepython
Compatibility
measured onNVIDIA A100
declared hardwaredeclared only
architectures—
dtypes

Benchmark evidence

8 measurements across 1 GPU, fastest first.

Operation / workload
Hardware
Latency
Rank
Observed
Fused linear + CPO lossbf16 · [2, 1024, 4096]
NVIDIA A100
14.5ms
#1 of 2
2024-11-14
Fused linear + CPO lossbf16 · [4, 1024, 4096]
NVIDIA A100
26.1ms
#1 of 2
2024-11-14
Fused linear + CPO lossbf16 · [2, 1024, 4096]
NVIDIA A100
39.5ms
#2 of 2
2024-11-14
Fused linear + CPO lossbf16 · [8, 1024, 4096]
NVIDIA A100
51.9ms
#1 of 2
2024-11-14
Fused linear + CPO lossbf16 · [4, 1024, 4096]
NVIDIA A100
77.0ms
#2 of 2
2024-11-14
Fused linear + CPO lossbf16 · [16, 1024, 4096]
NVIDIA A100
105.0ms
#1 of 2
2024-11-14
Fused linear + CPO lossbf16 · [8, 1024, 4096]
NVIDIA A100
154.3ms
#2 of 2
2024-11-14
Fused linear + CPO lossbf16 · [16, 1024, 4096]
NVIDIA A100
309.2ms
#2 of 2
2024-11-14

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:4a48076b1e37e0c4bef0af8dfd8a58ab3edf8b9a9d143e584ce235c14c7013ea
license declaredApache-2.0
license concludedApache-2.0
imported2026-08-20

Best evidence level for this revision: reported

JSON