ProductNVIDIA® L4 ADA GPU Computing Accelerator - 24GB GDDR6X - PCIe 4.0 x16 - Passive CoolingNVIDIA® L40S ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive CoolingNVIDIA® RTX A1000 - 8GB GDDR6 - PCIe 4.0 x8 - Active Cooling (4x mDP)NVIDIA® RTX PRO 4000 Blackwell - 24GB GDDR7 ECC - PCIe 5.0 x16 - Active Cooling (4xDP)NVIDIA® RTX PRO 4500 Blackwell - 32GB GDDR7 ECC - PCIe 5.0 x16 - Active Cooling (4xDP)NVIDIA® RTX PRO 5000 Blackwell - 48GB GDDR7 ECC - PCIe 5.0 x16 - Active Cooling (4xDP)NVIDIA® RTX PRO 6000 Blackwell Max-Q Workstation Edition - 96GB GDDR7 ECC - PCIe 5.0 x16 - Active Cooling (4xDP)
ActionSelectSelectSelectSelectSelectSelectSelect
Main Specifications
Product Series Nvidia L4Nvidia L40S
Core Type NVIDIA TENSORNVIDIA TENSOR
Core Clock Speed 795 MHz Base | 2040 MHz Boost
Host Interface PCI Express 4.0 x16PCI Express 4.0 x16PCI Express 4.0 x8PCI Express 5.0 x16PCI Express 5.0 x16PCI Express 5.0 x16PCI Express 5.0 x16
GPU Architecture Ada LovelaceAda Lovelace
Product Type WorkstationWorkstationWorkstationWorkstationWorkstation
Product Line NVIDIA Professional GraphicsNVIDIA Professional GraphicsNVIDIA Professional GraphicsNVIDIA Professional GraphicsNVIDIA Professional Graphics
Memory Technology GDDR6GDDR7GDDR7GDDR7GDDR7
Memory Capacity 8 GB24 GB with ECC32 GB with ECC48 GB with ECC96 GB with ECC
Max Displays 4 Displays4 Displays4 Displays4 Displays4 Displays
Detailed Specifications
Streaming Processor Cores 18,1762307 CUDA Cores8960 CUDA Parallel Processing Cores10,496 CUDA Parallel Processing Cores14,080 CUDA Parallel Processing Cores24,064 CUDA Parallel Processing Cores
NVIDIA Tensor Cores 568 | Gen 472280328440752
NVIDIA RT Cores 142 | Gen 3187082110188
Memory Clock Speed 6251 MHz
Memory Interface 192-bit128-bit192-bit256-bit384-bit512-bit
Max Memory Size 24 GB48GB GDDR6 with ECC
Max Memory Bandwidth 300 GB/s864 GB/s
ECC Protection On by Default
INT8 Tensor Core 485 TOPS | Sparsity733 teraFLOPS
TF32 Tensor Core 120 TFLOPS | Sparsity183 teraFLOPS
FP32 30.3 TFLOPS91.6 teraFLOPS
Peak BFLOAT16 Tensor Core 242 TFLOPS | Sparsity362.05 teraFLOPS
Peak FP16 Tensor Core 242 TFLOPS | Sparsity362.05 teraFLOPS
Peak FP8 Tensor Core 485 TFLOPS | Sparsity733 teraFLOPS
Peak INT4 Tensor Core 733 teraFLOPS
Total NVLink Bandwidth Not supported
Multi-Instance GPUs No
Tensor Performance 13.2 TFLOPS
NVENC | NVDEC 2 | 4 | 4 | JPEG Decoders | AV1 Encode and Decode3x l 3x (includes AV1 encode and decode)
Secure Boot with Root of Trust YesYes
NEBS Ready Yes | Level 3Level 3
Transistor Count 8.7 Billion45.6 Billion92.2 Billion92.2 Billion
DisplayPort Connectors None | vGPU Only4x DisplayPort 1.4a
Cooling PassivePassive
Dual Slot No
Dimensions 4.4" (H) x 10.5" (L)2.7" H x 6.4" L4.4" H x 9.5" L, single height4.4" H x 10.5" L4.4” H x 10.5” L, FHFL Dual Slot4.4” H x 10.5” L, FHFL Dual Slot
Form Factor 6.61” L x 2.71” H (Low-profile)
Lithography 8N | NVIDIA Custom Process4N NVIDIA Custom Process4N NVIDIA Custom Process4N NVIDIA Custom Process
Supplementary Power Connectors 1x 16-pin1x PCIe CEM5 16-pin1x PCIe CEM5 16-pin1x PCIe CEM5 16-pin1x PCIe CEM5 16-pin
Max Graphics Card Power (W) 72W350W50W140W200W300W300W
Processor AmpereNVIDIA Blackwell ArchitectureNVIDIA Blackwell ArchitectureNVIDIA Blackwell ArchitectureNVIDIA Blackwell Architecture
Memory Bandwidth 192 GB/sec672 GB/s896 GB/s1344 GB/s1792 GB/s
Peak Single Precision FP32 Performance 6.74 TFLOPS
NVLink Interconnect Not Supported
DisplayPort Output 4x DP 2.14x DP 2.14x DP 2.14x DP 2.1
Mini DisplayPort Output 4x mDisplayPort 1.4a
Minimum Recommended Power, Single Card (W) 550550600750
Minimum Recommended Power, 2-Way (W) 750
Minimum Recommended Power, 3-Way (W) 850
Minimum Recommended Power, 4-Way (W) 1000
Thermal Solution Active FanBlower Active FanBlower Active FanBlower Active FanBlower Active Fan
Slot Height Single Slot1-Slot2-Slot2-Slot2-Slot
ActionSelectSelectSelectSelectSelectSelectSelect