{{ 'fb_in_app_browser_popup.desc' | translate }} {{ 'fb_in_app_browser_popup.copy_link' | translate }}
{{ 'in_app_browser_popup.desc' | translate }}
NVIDIA, a leading brand in GPU technology, introduces the Nvidia L40S, designed specifically for high-performance computing. This GPU offers unparalleled GPU acceleration, making it ideal for advanced AI models, rendering tasks, and data center applications.
Not enough stock.
Your item was not added to your cart.
Not enough stock.
Please adjust your quantity.
{{'products.quick_cart.out_of_number_hint'| translate}}
{{'product.preorder_limit.hint'| translate}}
Limit {{ product.max_order_quantity }} per order.
Only {{ quantityOfStock }} item(s) left.
NVIDIA, a leading brand in GPU technology, introduces the Nvidia L40S, designed specifically for high-performance computing. This GPU offers unparalleled GPU acceleration, making it ideal for advanced AI models, rendering tasks, and data center applications. Engineers and procurement professionals seeking top-tier performance in their data centers will find the Nvidia L40S to be an excellent choice.
Feature | Nvidia L40S | Nvidia A100 | Nvidia V100 | AMD Instinct MI100 | Google TPU v4 | Intel Habana Gaudi | Graphcore IPU-M2000 |
---|---|---|---|---|---|---|---|
GPU Architecture | Ampere | Ampere | Volta | CDNA | TPU | Gaudi | Colossus Mk2 |
Memory | 48GB GDDR6 | 40GB HBM2 | 16GB HBM2 | 32GB HBM2 | 128GB | 32GB HBM2 | 450MB In-Processor Memory |
Memory Bandwidth | 1.4 TB/s | 1.6 TB/s | 900 GB/s | 1.23 TB/s | 600 GB/s | 900 GB/s | 30 TB/s |
Power Consumption | 300W | 400W | 250W | 300W | 280W | 200W | 400W |
Peak Performance | 24 TFLOPs | 312 TFLOPs | 120 TFLOPs | 184.6 TFLOPs | 100+ PFLOPs | 2 TAI TOPS | 64 IPU-Tiles |
Model | Description |
---|---|
Nvidia NVLink Bridge | High-speed interconnect for multiple GPUs. |
NVIDIA DGX Station A100 | Workgroup appliance for AI development. |
Quadro Sync II | Synchronize up to 32 displays or projectors. |
Nvidia L40S Specifications | |
GPU Architecture | Ampere |
Cores | 10752 CUDA Cores |
Tensor Cores | 336 Third-Generation Tensor Cores |
RT Cores | 84 Second-Generation RT Cores |
Memory | 48 GB GDDR6 |
Memory Interface | 384-bit |
Memory Bandwidth | 960 GB/s |
GPU Base Clock | 1230 MHz |
GPU Boost Clock | 1650 MHz |
Performance per Watt | High Efficiency |
Power Consumption | 300 W |
Interface | PCIe 4.0 x16 |
Display Connectors | 4x DisplayPort 1.4a |
Maximum Displays Supported | 4 |
Form Factor | Dual Slot |
Cooling | Active Cooling |
Operating Temperature | 0°C to 85°C |
Maximum Resolution | 7680x4320 @ 60Hz |
AI Performance | High-Performance AI Training and Inference |