Product	NVIDIA® A2 GPU Computing Accelerator - 16GB GDDR6 - PCIe 4.0 x8 - Passive Cooler (w/o CEC)	NVIDIA® A100 GPU Computing Accelerator - 80GB HBM2 - PCIe 4.0 x16 - Passive Cooling	NVIDIA® L4 ADA GPU Computing Accelerator - 24GB GDDR6X - PCIe 4.0 x16 - Passive Cooling	NVIDIA® L40 ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive Cooling	NVIDIA® L40S ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive Cooling	NVIDIA® H100 GPU Computing Accelerator - 80GB HBM2e - PCIe 5.0 x16 - Passive Cooling (w/o CEC)	NVIDIA® RTX A1000 - 8GB GDDR6 - PCIe 4.0 x8 - Active Cooling (4x mDP)	NVIDIA® RTX 4000 SFF Ada Generation - 20GB GDDR6 ECC - PCIe 4.0 x16 - Active Cooling (4x mDP)
Action	Select	Select	Select	Select	Select	Select	Select	Select
Main Specifications
Product Series	Nvidia A2	Nvidia A100	Nvidia L4	Nvidia L40	Nvidia L40S	Nvidia H100
Core Type	NVIDIA TENSOR	NVIDIA TENSOR	NVIDIA TENSOR	NVIDIA TENSOR	NVIDIA TENSOR	NVIDIA TENSOR
Core Clock Speed	1440 MHz (1770 MHz Boost Clock)		795 MHz Base | 2040 MHz Boost
Host Interface	PCI Express 4.0 x8	PCI Express 4.0 x16	PCI Express 4.0 x16	PCI Express 4.0 x16	PCI Express 4.0 x16	PCI Express 5.0 x16	PCI Express 4.0 x8	PCI Express 4.0 x16
GPU Architecture	Ampere	Ampere	Ada Lovelace	Ada Lovelace	Ada Lovelace	Hopper
Product Type							Workstation	Workstation
Product Line							NVIDIA Professional Graphics	NVIDIA Professional Graphics
Memory Technology							GDDR6	GDDR6
Memory Capacity							8 GB	20 GB GDDR6 ECC
Max Displays							4 Displays
Detailed Specifications
Streaming Processor Cores	1280 CUDA Cores				18,176		2307 CUDA Cores	6144 CUDA Cores
NVIDIA Tensor Cores	40 | Gen 3				568 | Gen 4		72	192 | Gen 4
NVIDIA RT Cores	10 | Gen 2				142 | Gen 3		18	48 | Gen 3
PCIe x16 Interconnect Bandwidth		PCIe Gen4 64 GB/s				PCIe Gen5: 128GB/s
Memory Clock Speed	6251 MHz		6251 MHz
Memory Interface	128-bit		192-bit				128-bit	160-bit
Max Memory Size	16 GB GDDR6 ECC	80 GB	24 GB	48 GB GDDR6 with ECC	48GB GDDR6 with ECC	80 GB
Max Memory Bandwidth	200 GB/s	1,935 GB/s	300 GB/s		864 GB/s	2.0TB/s
ECC Protection	On by Default		On by Default
Peak FP64		9.7 TFLOPS				24 teraFLOPS
Peak FP64 Tensor Core		19.5 TFLOPS				48 teraFLOPS
INT8 Tensor Core		624 TOPS	485 TOPS | Sparsity		733 teraFLOPS	3,200 TOPS
TF32 Tensor Core	9 TFLOPS | 18 TFLOPS Sparsity	156 TFLOPS	120 TFLOPS | Sparsity		183 teraFLOPS	800 teraFLOPS
FP32	4.5 TFLOPS	19.5 TFLOPS	30.3 TFLOPS		91.6 teraFLOPS	48 teraFLOPS
Peak BFLOAT16 Tensor Core		312 TFLOPS	242 TFLOPS | Sparsity		362.05 teraFLOPS	1,600 teraFLOPS
Peak FP16 Tensor Core	18 TFLOPS | 36 TFLOPS Sparsity	312 TFLOPS	242 TFLOPS | Sparsity		362.05 teraFLOPS	1,600 teraFLOPS
Peak FP8 Tensor Core			485 TFLOPS | Sparsity		733 teraFLOPS	3,200 teraFLOPS
Peak INT4 Tensor Core					733 teraFLOPS
Total NVLink Bandwidth		600 GB/s (via NVLink Bridge for up to 2-GPUs)			Not supported	600GB/s
Multi-Instance GPUs		7 MIGs at 10GB			No	Up to 7 MIGS @ 10GB each
Tensor Performance							13.2 TFLOPS	306.8 TFLOPS
NVIDIA CUDA™ Technology	11.1 or later							Yes
vGPU Software Support				NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation (vWS)
NVENC | NVDEC			2 | 4 | 4 | JPEG Decoders | AV1 Encode and Decode	3x | 3x (Includes AV1 Encode & Decode)	3x l 3x (includes AV1 encode and decode)
Secure Boot with Root of Trust			Yes	Yes	Yes
NEBS Ready			Yes | Level 3	Yes / Level 3	Level 3
Peak INT4 Performance	72 TOPS | 144 TOPS Sparsity
Peak INT8 Performance	36 TOPS | 72 TOPS Sparsity
Transistor Count							8.7 Billion	35.8 Billion
DisplayPort Connectors			None | vGPU Only	4x DP 1.4a	4x DisplayPort 1.4a
Cooling	Passive	Passive	Passive	Passive	Passive	Passive
Dual Slot	Single-slot	Yes	No	Yes		Yes
Dimensions	6.61” L x 2.71” H			4.4" (H) x 10.5" (L)	4.4" (H) x 10.5" (L)		2.7" H x 6.4" L	2.7” H x 6.6”L
Form Factor	Low-Profile PCIe		6.61” L x 2.71” H (Low-profile)	PCIe		PCIe
Lithography							8N | NVIDIA Custom Process
Supplementary Power Connectors		1x 8-pin CPU (EPS12V)		1x 16-pin PCIe CEM5	1x 16-pin	1x 16-pin PCIe CEM5		No Auxiliary Power Required
Max Graphics Card Power (W)	40-60 W | Configurable	300W	72W	300W	350W	350W	50W	70W
Processor							Ampere	NVIDIA Ada Lovelace
Memory Bandwidth							192 GB/sec	320 GB/s
Peak Single Precision FP32 Performance							6.74 TFLOPS	19.2 TFLOPS
NVLink Interconnect							Not Supported
RT Core Performance								44.3 TFLOPS
VR Ready								Yes
Mini DisplayPort Output							4x mDisplayPort 1.4a	4x mDP 1.4a
Thermal Solution							Active Fan	Active Heatsink
Slot Height							Single Slot	Low Profile Dual Slot
Action	Select	Select	Select	Select	Select	Select	Select	Select