QuData

Rent L40S by the Hour by QuData Ai

Explore the specifications and benchmarks of the L40S. Compare hardware configurations and performance metrics to optimize your high-performance AI and ML workloads with enhanced GPU computing architecture.

GPU Description

Explore the specifications and benchmarks of the L40S. Compare hardware configurations and performance metrics to optimize your high-performance AI and ML workloads with enhanced GPU computing architecture.

GPU Characteristics

GPU typeL40S
GPU connectionPCIe
GPU model nameNVIDIA L40S
Driver version535.183.01
GPU memory48GB GDDR6
Power Limit (W)350.0
GPU temperature (°C)62
GPU Clock Speed (MHz)2520
Memory Clock Speed (MHz)9001
Pstate0

GPU Configuration

CUDA cores18176
Tensor core568 (4th generation)
Memory bandwidth864 GB/s
Support for FP16/INT8Yes
ArchitectureAda Lovelace
VirtualizationNVIDIA vGPU

Benchmarks

ffmpeg~9ms
Coremark (iterations per second)~80000-90000
Llama2 Inference (tokens per second)~155-185
Tensorflow Mixed Training~4.1-4.8
Stable Diffusion (images/min)~45-55
Video Encoding (4K)~95-115 fps

Providers with L40S