QuData

Rent A40 by the Hour by QuData Ai

Explore the specifications and benchmarks of the A40. Compare hardware configurations and performance metrics to optimize your professional AI and graphics workloads with a proven GPU computing architecture.

GPU Description

Explore the specifications and benchmarks of the A40. Compare hardware configurations and performance metrics to optimize your professional AI and graphics workloads with a proven GPU computing architecture.

GPU Characteristics

GPU typeA40
GPU connectionPCIe
GPU model nameNVIDIA A40
Driver version535.183.01
GPU memory48GB GDDR6
Power Limit (W)300.0
GPU temperature (°C)55
GPU Clock Speed (MHz)1740
Memory Clock Speed (MHz)7251
Memory bandwidth696 GB/s

GPU Configuration

CUDA cores10752
Tensor core336 (3rd generation)
Pstate0
Support for FP16/INT8Yes
ArchitectureAmpere
VirtualizationNVIDIA vGPU

Benchmarks

ffmpeg~18ms
Coremark (iterations per second)~55000-65000
Llama2 Inference (tokens per second)~95-125
Tensorflow Mixed Training~2.8-3.5
Stable Diffusion (images/min)~28-38
Video Encoding (4K)~65-85 fps

Providers with A40