ProductNVIDIA® L40 ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive CoolingNVIDIA® L40S ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive CoolingNVIDIA® H100 NVL GPU Computing Accelerator - 94GB HBM3 - PCIe 5.0 x16 - Passive CoolingNVIDIA® RTX A4000 - 16GB GDDR6 - PCIe 4.0 x16 - Active Cooling (4xDP)NVIDIA® RTX A6000 - 48GB GDDR6 - PCIe 4.0 x16 - Active Cooling (4xDP)
ActionSelectSelectSelectSelectSelect
Main Specifications
Product Series Nvidia L40Nvidia L40SNvidia H100 NVL
Core Type NVIDIA TENSORNVIDIA TENSORNVIDIA TENSOR
Host Interface PCI Express 4.0 x16PCI Express 4.0 x16PCI Express 5.0 x16PCI Express 4.0 x16PCI Express 4.0 x16
GPU Architecture Ada LovelaceAda LovelaceHopper
Product Type WorkstationWorkstation
Product Line NVIDIA Professional GraphicsNVIDIA Professional Graphics
Memory Technology GDDR6 with ECCGDDR6
Memory Capacity 16 GB GDDR6 with ECC48 GB
Max Displays 4 Displays
Detailed Specifications
Streaming Processor Cores 18,1766144 CUDA Cores10752 Shading Units
NVIDIA Tensor Cores 568 | Gen 4192336
NVIDIA RT Cores 142 | Gen 34884
PCIe x16 Interconnect Bandwidth PCIe Gen5: 128GB/s
Memory Clock Speed 2000 MHz 16 Gbps effective
Memory Interface 256-bit384-bit
Max Memory Size 48 GB GDDR6 with ECC48GB GDDR6 with ECC94 GB
Max Memory Bandwidth 864 GB/s7.8TB/s
Peak FP64 68 teraFLOPs
Peak FP64 Tensor Core 134 teraFLOPs
INT8 Tensor Core 733 teraFLOPS7,916 TOPS
TF32 Tensor Core 183 teraFLOPS1,979 teraFLOPs
FP32 91.6 teraFLOPS134 teraFLOPs
Peak BFLOAT16 Tensor Core 362.05 teraFLOPS3,958 teraFLOPs
Peak FP16 Tensor Core 362.05 teraFLOPS3,958 teraFLOPs
Peak FP8 Tensor Core 733 teraFLOPS7,916 teraFLOPs
Peak INT4 Tensor Core 733 teraFLOPS
Total NVLink Bandwidth Not supported600GB/s
Multi-Instance GPUs No
NVIDIA CUDA™ Technology YesYes
vGPU Software Support NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation (vWS)
NVENC | NVDEC 3x | 3x (Includes AV1 Encode & Decode)3x l 3x (includes AV1 encode and decode)
Secure Boot with Root of Trust YesYes
NEBS Ready Yes / Level 3Level 3
Transistor Count 17.4 Billion28.3 Billion
DisplayPort Connectors 4x DP 1.4a4x DisplayPort 1.4a
Cooling PassivePassivePassive
Dual Slot YesYes
Dimensions 4.4" (H) x 10.5" (L)4.4" (H) x 10.5" (L)4.4” H x 9.5” L4.4" (H) x 10.5" (L)
Form Factor PCIePCIe
Lithography 8nmSamsung 8nm
Supplementary Power Connectors 1x 16-pin PCIe CEM51x 16-pin1x 6-pin PCIe1x 8-pin EPS
Max Graphics Card Power (W) 300W350W400W140W300W
Processor Ampere (GA104)Ampere (GA102)
Memory Bandwidth 448 GB/sec768 GB/s
Core Clock Speed 1455 MHz Base Clock
1860 MHz Boost Clock
L2 Cache Size 6 MB
API Support CUDA 8.5, OpenCL 2.0
Shader Model 6.5, OpenGL 4.6, DirectX 12 Ultimate (12_2), Vulkan 1.2
Texture Fill Rate 625 GTexel/s
Graphics Resolution Max Digital Resolution: 7680 x 4320 x36 bpp at 60 Hz7680 x 4320 x36 bpp at 60 Hz
Peak Double Precision FP64 Performance 1,250 GFLOPS (1:32)
Peak Single Precision FP32 Performance 38.7 TFLOPS
Peak Half Precision FP16 Performance 40.00 TFLOPS (1:1)
Deep Learning TFLOPS 153.4 TFLOPS
Multi-GPU Scalability NVLINK 2-way low profile (2-slot and 3-slot bridges) connects 2x NVIDIA RTX A6000
NVLink Interconnect 112.5 GB/s (bidirectional)
VR Ready Yes
Vulkan API 1.2
DisplayPort Output 4x DisplayPort 1.4a4x DisplayPort 1.4a
Minimum Recommended Power, Single Card (W) 300W700W
Minimum Recommended Power, 2-Way (W) 500850
Minimum Recommended Power, 3-Way (W) 8501000
Minimum Recommended Power, 4-Way (W) 10001200
Thermal Solution Active HeatsinkActive Heatsink
Slot Height Single Slot2-Slot
ActionSelectSelectSelectSelectSelect