Ampere Architecture
Memory size 80GB HBM2e with ECC
Memory bandwidth 1,935GB/s
FP64 - 9.7 TFLOPS
FP64 Tensor Core - 19.5 TFLOPS
FP32 - 19.5 TFLOPS
Tensor Float 32 (TF32) - 156 TFLOPS | 312 TFLOPS
BFLOAT16 Tensor Core - 312 TFLOPS | 624 TFLOPS
FP16 Tensor Core - 312 TFLOPS | 624 TFLOPS
INT8 Tensor Core - 624 TOPS | 1248 TOPS*
Interconnect NVIDIA® NVLink® Bridge
for 2 GPU: 600GB/s PCIe 4.0 x16 (64GB/s)
Max Thermal Design Power (TDP) 300W
Server Options Partner and
NVIDIA-Certified Systems™ with 1-8 GPU
Ampere Architecture
Memory Size 40 GB HBM2 with ECC
Memory Bus width 5120-bit
Memory Bandwidth 1,6 TB/s
Cuda Cores 6,912
9.7 Tflops (GPU Boost Clocks)
19.5 Tflops (GPU Boost Clocks)
312Tflops (GPU Boost Clocks)
System interface PCI Express 4.0 x16
Max power consumption 250 W
Ampere Architecture
GPU Memory 64 GB GDDR6
with Error Correcting Code (ECC)
GPU Memory Bandwidth | 4x 232 GB/s
Max Power Consumption 250 W
Interconnect PCI Express Gen 4.0 x16
Thermal Solution Passive
vGPU Software Support
NVIDIA Virtual PC (vPC)
NVIDIA Virtual Applications (vApps)
NVIDIA RTX Workstation (vWS)
NVIDIA Virtual Compute Server (vCS)
Ampere Architecture
16 GB GDDR6 with ECC
48 RT Cores / 192 Tensor Cores / 6,144 CUDA Cores
PCI Express 4.0 x16
SP(FP32) 19.2 TFLOPS
RT Core Performance 37.4 TFLOPS
Tensor Performance 153.4 TFLOPS
4x DisplayPort 1.4
Memory Bandwidth Up to 448 GB/s
Memory Interface 256 bit
Ampere Architecture
FP32 31.2 TFLOPS
TF32 Tensor Core 62.5 TFLOPS | 125 TFLOPS*
BFLOAT16 Tensor Core 126 TFLOPS | 250 TFLOPS*
FP16 Tensor Core 165 TFLOPS | 330 TF*
INT8 Tensor Core 250 TOPS | 500 TOPS*
INT4 Tensor Core 500 TOPS | 1000 TOPS*
GPU Memory 24 GB GDDR6
Memory Bandwidth 600 GB/s
Thermal Solutions Passive
Maximum Power Consumption 150 W
System Interface PCIe Gen 4.0 | 64 GB/s
vGPU Support Yes
Ampere Architecture
RT Cores 10
Peak FP32 4.5 TF
TF32 Tensor Core 9 TF | 18 TF¹
BFLOAT16 TensorCore 18 TF | 36 TF
Peak FP16 Tensor Core 18 TF | 36 TF
Peak INT8 Tensor Core 36 TOPS | 72 TOPS
Peak INT4 Tensor Core 72 TOPS | 144 TOPS
Media engines 1 video encoder
2 video decoders (includes AV1 decode)
GPU memory 16GB GDDR6
GPU memory bandwidth 200GB/s
Max thermal design power(TDP) 40-60W
Interconnect PCIe Gen4 x8
vGPU Support Yes
Turing Architecture
NVIDIA Turing Tensor Cores 320
NVIDIA CUDA® Cores 2,560
Single-Precision 8.1 TFLOPS
Mixed-Precision (FP16/FP32) 65 TFLOPS
INT8 130 TOPS
INT4 260 TOPS
GPU Memory 16GB GDDR6 300 GB/s