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® H100 GPU Computing Accelerator - 80GB HBM2e - PCIe 5.0 x16 - Passive Cooling (w/o CEC)
ActionSelectSelectSelect
Main Specifications
Product Series Nvidia L4Nvidia L40SNvidia H100
Core Type NVIDIA TENSORNVIDIA TENSORNVIDIA TENSOR
Core Clock Speed 795 MHz Base | 2040 MHz Boost
Host Interface PCI Express 4.0 x16PCI Express 4.0 x16PCI Express 5.0 x16
GPU Architecture Ada LovelaceAda LovelaceHopper
Detailed Specifications
Streaming Processor Cores 18,176
NVIDIA Tensor Cores 568 | Gen 4
NVIDIA RT Cores 142 | Gen 3
PCIe x16 Interconnect Bandwidth PCIe Gen5: 128GB/s
Memory Clock Speed 6251 MHz
Memory Interface 192-bit
Max Memory Size 24 GB48GB GDDR6 with ECC80 GB
Max Memory Bandwidth 300 GB/s864 GB/s2.0TB/s
ECC Protection On by Default
Peak FP64 24 teraFLOPS
Peak FP64 Tensor Core 48 teraFLOPS
INT8 Tensor Core 485 TOPS | Sparsity733 teraFLOPS3,200 TOPS
TF32 Tensor Core 120 TFLOPS | Sparsity183 teraFLOPS800 teraFLOPS
FP32 30.3 TFLOPS91.6 teraFLOPS48 teraFLOPS
Peak BFLOAT16 Tensor Core 242 TFLOPS | Sparsity362.05 teraFLOPS1,600 teraFLOPS
Peak FP16 Tensor Core 242 TFLOPS | Sparsity362.05 teraFLOPS1,600 teraFLOPS
Peak FP8 Tensor Core 485 TFLOPS | Sparsity733 teraFLOPS3,200 teraFLOPS
Peak INT4 Tensor Core 733 teraFLOPS
Total NVLink Bandwidth Not supported600GB/s
Multi-Instance GPUs NoUp to 7 MIGS @ 10GB each
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
DisplayPort Connectors None | vGPU Only4x DisplayPort 1.4a
Cooling PassivePassivePassive
Dual Slot NoYes
Dimensions 4.4" (H) x 10.5" (L)
Form Factor 6.61” L x 2.71” H (Low-profile)PCIe
Supplementary Power Connectors 1x 16-pin1x 16-pin PCIe CEM5
Max Graphics Card Power (W) 72W350W350W
ActionSelectSelectSelect