fp16 ln
Eight independent scalar f16 ln-add chains per thread, unrolled 4x. Same shape as f32-log but entirely in half precision.
Measured in FLOP/s. Higher values mean better performance. Benchmark your device.
Performance distribution
379 MFLOP/s74.3 GFLOP/s14.6 TFLOP/s
379 MFLOP/s74.3 GFLOP/s14.6 TFLOP/s
Performance thresholds
99% of results achieve at least 713 MFLOP/s
95% of results achieve at least 20.6 GFLOP/s
90% of results achieve at least 48.8 GFLOP/s
Median (50th percentile) 763 GFLOP/s
Best performer
Highest recorded performance: 14.6 TFLOP/s. View benchmark profile.
Recorded 9/14/2026, 2:52:24 PM
WebGPU device
Vendornvidia
Device—
Architectureblackwell
Description—
WebGL device
VendorGoogle Inc. (NVIDIA)
RendererANGLE (NVIDIA, NVIDIA GeForce RTX 5090 (0x00002B85) Direct3D11 vs_5_0 ps_5_0, D3D11)
Browser
NameChrome
Version153.0.0.0
Operating system
NameWindows
Version10