| | | | |
Product | NVIDIA® L4 ADA GPU Computing Accelerator - 24GB GDDR6X - PCIe 4.0 x16 - Passive Cooling | NVIDIA® L40S ADA GPU Computing Accelerator - 48GB GDDR6 - PCIe 4.0 x16 - Passive Cooling | NVIDIA® H100 NVL GPU Computing Accelerator - 94GB HBM3 - PCIe 5.0 x16 - Passive Cooling | NVIDIA® RTX 4000 Ada Generation - 20GB GDDR6 ECC - PCIe 4.0 x16 - Active Cooling (4xDP) |
Action | Select | Select | Select | Select |
Main Specifications |
Product Series |
Nvidia L4 | Nvidia L40S | Nvidia H100 NVL | |
Core Type |
NVIDIA TENSOR | NVIDIA TENSOR | NVIDIA TENSOR | |
Core Clock Speed |
795 MHz Base | 2040 MHz Boost | | | |
Host Interface |
PCI Express 4.0 x16 | PCI Express 4.0 x16 | PCI Express 5.0 x16 | PCI Express 4.0 x16 |
GPU Architecture |
Ada Lovelace | Ada Lovelace | Hopper | |
Product Line |
| | | NVIDIA Professional Graphics |
Memory Technology |
| | | GDDR6 |
Memory Capacity |
| | | 20 GB GDDR6 ECC |
Detailed Specifications |
Streaming Processor Cores |
| 18,176 | | 6144 CUDA Cores |
NVIDIA Tensor Cores |
| 568 | Gen 4 | | 192 |
NVIDIA RT Cores |
| 142 | Gen 3 | | 48 |
PCIe x16 Interconnect Bandwidth |
| | PCIe Gen5: 128GB/s | |
Memory Clock Speed |
6251 MHz | | | |
Memory Interface |
192-bit | | | 160-bit |
Max Memory Size |
24 GB | 48GB GDDR6 with ECC | 94 GB | |
Max Memory Bandwidth |
300 GB/s | 864 GB/s | 7.8TB/s | |
ECC Protection |
On by Default | | | |
Peak FP64 |
| | 68 teraFLOPs | |
Peak FP64 Tensor Core |
| | 134 teraFLOPs | |
INT8 Tensor Core |
485 TOPS | Sparsity | 733 teraFLOPS | 7,916 TOPS | |
TF32 Tensor Core |
120 TFLOPS | Sparsity | 183 teraFLOPS | 1,979 teraFLOPs | |
FP32 |
30.3 TFLOPS | 91.6 teraFLOPS | 134 teraFLOPs | |
Peak BFLOAT16 Tensor Core |
242 TFLOPS | Sparsity | 362.05 teraFLOPS | 3,958 teraFLOPs | |
Peak FP16 Tensor Core |
242 TFLOPS | Sparsity | 362.05 teraFLOPS | 3,958 teraFLOPs | |
Peak FP8 Tensor Core |
485 TFLOPS | Sparsity | 733 teraFLOPS | 7,916 teraFLOPs | |
Peak INT4 Tensor Core |
| 733 teraFLOPS | | |
Total NVLink Bandwidth |
| Not supported | 600GB/s | |
Multi-Instance GPUs |
| No | | |
Tensor Performance |
| | | 327.6 TFLOPS |
NVENC | NVDEC |
2 | 4 | 4 | JPEG Decoders | AV1 Encode and Decode | 3x l 3x (includes AV1 encode and decode) | | |
Secure Boot with Root of Trust |
Yes | Yes | | |
NEBS Ready |
Yes | Level 3 | Level 3 | | |
Transistor Count |
| | | 35.8 Billion |
DisplayPort Connectors |
None | vGPU Only | 4x DisplayPort 1.4a | | |
Cooling |
Passive | Passive | Passive | |
Dual Slot |
No | | Yes | |
Dimensions |
| 4.4" (H) x 10.5" (L) | | 4.4" (H) x 9.5"(L) |
Form Factor |
6.61” L x 2.71” H (Low-profile) | | PCIe | |
Supplementary Power Connectors |
| 1x 16-pin | | |
Max Graphics Card Power (W) |
72W | 350W | 400W | 130W |
Processor |
| | | NVIDIA Ada Lovelace |
Memory Bandwidth |
| | | 360 GB/s |
Peak Single Precision FP32 Performance |
| | | 26.7 TFLOPS |
NVLink Interconnect |
| | | Not Supported |
RT Core Performance |
| | | 61.8 TFLOPS |
Mini DisplayPort Output |
| | | 4x mDP 1.4a |
Thermal Solution |
| | | Blower Active Fan |
Slot Height |
| | | Single Slot |
Action | Select | Select | Select | Select |