Skip to content
‹ All benchmarks

NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition

2026-09-07 · 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 RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)015304560124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 49.8NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 49.1NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 42.1NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 38.1NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 35.9
HoistingOn

CT × CT matrix multiplication

64 × 64 · Entry depth 0
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)08.7517.526.335124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 28.3NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 26.3NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 22.9NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 21NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 19.7
HoistingOn

Nonlinear polynomial evaluation

Entry depth 0
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)010203040124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 34NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 33.3NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 28.7NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 23.4NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 22.1

NTT

Entry depth 0
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)06,25012,50018,80025,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 14,400NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 17,600NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 18,300NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 13,200NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 9,600

Key switching

Entry depth 0
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)02505007501,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 886NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 870NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 780NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 621NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 575

RNS arithmetic

Entry depth 0
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)025,00050,00075,000100,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 61,600NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 87,300NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 42,800NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 33,700NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 34,100

Matrix operations

8 operations

Rotate (1 offset)

Entry depth 0 · Shift +1
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)0225450675900124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 775NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 758NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 661NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 549NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 488

Rotate-many (7 offsets)

HoistingOn
Entry depth 0 · Shifts +1…+7 · one input
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)062.5125188250124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 203NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 181NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 139NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 127NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 122

Multiply (PT × CT)

Entry depth 0 · No rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)011,30022,50033,80045,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 38,500NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 13,200NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 10,100NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 10,100NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 10,100

Sum products (8 PT × CT terms)

Entry depth 0 · Sum of eight products · no rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)01,7503,5005,2507,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 6,020NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 5,480NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 5,170NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 5,140NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 5,040

Rescale (2 components)

Entry depth 0 · One Q depth group
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)01,5003,0004,5006,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,790NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 5,190NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,840NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 3,530NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 3,300

Multiply (CT × CT)

Entry depth 0 · Distinct inputs · no relinearization or rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)08,75017,50026,30035,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 29,700NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 15,600NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 13,900NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 14,200NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 14,300

Add (CT + CT)

Entry depth 0 · Same depth and scale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)012,50025,00037,50050,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 40,400NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 19,700NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 16,600NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 16,900NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 17,100

Relinearize

Entry depth 0 · 3 → 2 components · no rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)0225450675900124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 781NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 770NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 676NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 558NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 498

Polynomial operations

Additional operations

Square (CT)

Entry depth 0 · Same input · no relinearization or rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)08,75017,50026,30035,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 29,900NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 26,300NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 18,500NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 19,000NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 19,400

Multiply scalar & rescale

Entry depth 0 · Scalar −0.3 · one rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)01,2502,5003,7505,000124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,430NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,170NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,110NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 3,100NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 2,930

Advance depth

Entry depth 0 · Multiply by one & rescale
NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Throughput (tasks/s)01,3802,7504,1305,500124816Batch sizeNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,500NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,200NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 4,110NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 3,100NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition · 2,960
Measurement conditions
HardwareNVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
Host CPUAMD Ryzen Threadripper PRO 9965WX 24-Cores
Intra-op / inter-op threads24 / 24
PyTorch / CUDA2.13.0+cu130 / 13.0
ExecutionEager CKKS and Backend RNS/NTT · prepared inputs and resources
Additional measurements · 2026-09-08325 cases · 20260908T203545Z-67c10251
Additional measurements · 2026-09-09245 cases · 20260909T144301Z-22862357