Skip to content
KernelIndex
Search⌘K

Transformers

Hugging Face Transformers · python · Apache-2.0
holds 2 records

Use it

Source available upstream · Apache-2.0View source →

No install recipe recorded for this revision.

interfacepython
Compatibility
measured onNVIDIA H100
declared hardwaredeclared only
architectures—
dtypes

Benchmark evidence

18 measurements across 1 GPU, fastest first.

Operation / workload
Hardware
Latency
Rank
Observed
Fused add + RMSNormfp32 · [2048, 1024]
NVIDIA H100
45.5µs
#2 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 2048]
NVIDIA H100
72.2µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 1024]
NVIDIA H100
113.6µs
#1 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 4096]
NVIDIA H100
149.1µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 2048]
NVIDIA H100
209.4µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 1024]
NVIDIA H100
229.6µs
#1 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 8192]
NVIDIA H100
271.1µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 2048]
NVIDIA H100
281.9µs
#2 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 4096]
NVIDIA H100
392.7µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 16384]
NVIDIA H100
509.2µs
#3 of 3
2026-04-07
Show all 18 measurements ›
Fused add + RMSNormfp32 · [2048, 4096]
NVIDIA H100
540.4µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 8192]
NVIDIA H100
720.7µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 32768]
NVIDIA H100
982.8µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 8192]
NVIDIA H100
989.8µs
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 16384]
NVIDIA H100
1.37ms
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 16384]
NVIDIA H100
1.87ms
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 32768]
NVIDIA H100
2.65ms
#3 of 3
2026-04-07
Fused add + RMSNormfp32 · [2048, 32768]
NVIDIA H100
3.63ms
#3 of 3
2026-04-07

Reported · How evidence levels are derived →

Source and license

sourceavailable
revision digestsha256:8a6a941077b6fb0ddcc8d75d257d1738f16cdc13ef8d4073ed447bebc48c807c
license declaredApache-2.0
license concludedApache-2.0
imported2026-08-18

Best evidence level for this revision: reported

JSON