Skip to content
‹ All benchmarks

2026-09-08 · Linux · PyTorch 2.13.0+cu130

N = 65,536 · Scale 250

ScalingThroughput
Log Y

Workloads and operations

PT × CT matrix multiplication

64 × 64 · Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)02.384.757.139.5124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 8.07NVIDIA GeForce RTX 4060 Ti · 7.09NVIDIA GeForce RTX 4060 Ti · 6.73NVIDIA GeForce RTX 4060 Ti · 6.71
HoistingOn

CT × CT matrix multiplication

64 × 64 · Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)01.252.53.755124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 4.36NVIDIA GeForce RTX 4060 Ti · 3.9NVIDIA GeForce RTX 4060 Ti · 3.7
HoistingOn

Nonlinear polynomial evaluation

Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)01.633.254.886.5124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 5.51NVIDIA GeForce RTX 4060 Ti · 4.26NVIDIA GeForce RTX 4060 Ti · 4.08NVIDIA GeForce RTX 4060 Ti · 4.1NVIDIA GeForce RTX 4060 Ti · 4.12

NTT

Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)08751,7502,6303,500124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 2,740NVIDIA GeForce RTX 4060 Ti · 1,960NVIDIA GeForce RTX 4060 Ti · 1,580NVIDIA GeForce RTX 4060 Ti · 1,620NVIDIA GeForce RTX 4060 Ti · 1,630

Key switching

Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)037.575113150124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 129NVIDIA GeForce RTX 4060 Ti · 115NVIDIA GeForce RTX 4060 Ti · 107NVIDIA GeForce RTX 4060 Ti · 107NVIDIA GeForce RTX 4060 Ti · 107

RNS arithmetic

Entry depth 0
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)01,7503,5005,2507,000124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 5,710NVIDIA GeForce RTX 4060 Ti · 5,460NVIDIA GeForce RTX 4060 Ti · 5,600NVIDIA GeForce RTX 4060 Ti · 5,690NVIDIA GeForce RTX 4060 Ti · 5,920

Matrix operations

8 operations

Rotate (1 offset)

Entry depth 0 · Shift +1
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)037.575113150124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 116NVIDIA GeForce RTX 4060 Ti · 103NVIDIA GeForce RTX 4060 Ti · 94.9NVIDIA GeForce RTX 4060 Ti · 94.8NVIDIA GeForce RTX 4060 Ti · 94.9

Rotate-many (7 offsets)

HoistingOn
Entry depth 0 · Shifts +1…+7 · one input
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)07.51522.530124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 24.7NVIDIA GeForce RTX 4060 Ti · 22NVIDIA GeForce RTX 4060 Ti · 21.5NVIDIA GeForce RTX 4060 Ti · 21.3NVIDIA GeForce RTX 4060 Ti · 21.3

Multiply (PT × CT)

Entry depth 0 · No rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)07501,5002,2503,000124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 2,660NVIDIA GeForce RTX 4060 Ti · 1,680NVIDIA GeForce RTX 4060 Ti · 1,700NVIDIA GeForce RTX 4060 Ti · 1,720NVIDIA GeForce RTX 4060 Ti · 1,710

Sum products (8 PT × CT terms)

Entry depth 0 · Sum of eight products · no rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)03757501,1301,500124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 870NVIDIA GeForce RTX 4060 Ti · 892NVIDIA GeForce RTX 4060 Ti · 903NVIDIA GeForce RTX 4060 Ti · 908NVIDIA GeForce RTX 4060 Ti · 910

Rescale (2 components)

Entry depth 0 · One Q depth group
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)0225450675900124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 799NVIDIA GeForce RTX 4060 Ti · 578NVIDIA GeForce RTX 4060 Ti · 586NVIDIA GeForce RTX 4060 Ti · 590NVIDIA GeForce RTX 4060 Ti · 594

Multiply (CT × CT)

Entry depth 0 · Distinct inputs · no relinearization or rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)07501,5002,2503,000124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 2,200NVIDIA GeForce RTX 4060 Ti · 2,310NVIDIA GeForce RTX 4060 Ti · 2,370NVIDIA GeForce RTX 4060 Ti · 2,400NVIDIA GeForce RTX 4060 Ti · 2,430

Add (CT + CT)

Entry depth 0 · Same depth and scale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)08751,7502,6303,500124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 2,620NVIDIA GeForce RTX 4060 Ti · 2,730NVIDIA GeForce RTX 4060 Ti · 2,810NVIDIA GeForce RTX 4060 Ti · 2,920NVIDIA GeForce RTX 4060 Ti · 2,960

Relinearize

Entry depth 0 · 3 → 2 components · no rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)037.575113150124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 121NVIDIA GeForce RTX 4060 Ti · 106NVIDIA GeForce RTX 4060 Ti · 97.6NVIDIA GeForce RTX 4060 Ti · 97.4NVIDIA GeForce RTX 4060 Ti · 97.4

Polynomial operations

Additional operations

Square (CT)

Entry depth 0 · Same input · no relinearization or rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)01,0002,0003,0004,000124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 3,000NVIDIA GeForce RTX 4060 Ti · 3,160NVIDIA GeForce RTX 4060 Ti · 3,300NVIDIA GeForce RTX 4060 Ti · 3,340NVIDIA GeForce RTX 4060 Ti · 3,350

Multiply scalar & rescale

Entry depth 0 · Scalar −0.3 · one rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)0188375563750124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 639NVIDIA GeForce RTX 4060 Ti · 486NVIDIA GeForce RTX 4060 Ti · 489NVIDIA GeForce RTX 4060 Ti · 491NVIDIA GeForce RTX 4060 Ti · 493

Advance depth

Entry depth 0 · Multiply by one & rescale
NVIDIA GeForce RTX 4060 Ti
Throughput (tasks/s)0188375563750124816Batch sizeNVIDIA GeForce RTX 4060 Ti · 639NVIDIA GeForce RTX 4060 Ti · 486NVIDIA GeForce RTX 4060 Ti · 489NVIDIA GeForce RTX 4060 Ti · 491NVIDIA GeForce RTX 4060 Ti · 494
Measurement conditions
HardwareNVIDIA GeForce RTX 4060 Ti
Host CPUAMD EPYC 7302P 16-Core Processor
Intra-op / inter-op threads16 / 32
PyTorch / CUDA2.13.0+cu130 / 13.0
ExecutionEager CKKS and Backend RNS/NTT · prepared inputs and resources
Additional measurements · 2026-09-09245 cases · 20260909T154817Z-7547636c