All GPUs

NVIDIA · Ada Lovelace

NVIDIA RTX 4090

A popular, cost-effective card for inference and image generation.

  • On-demand
  • 2 years
  • 3 years
Request a quote

Specifications

Architecture
Ada Lovelace
GPU memory
24 GB GDDR6X
Memory bandwidth
1 TB/s
Interconnect
PCIe 4.0
Form factor
PCIe

Best for

  • Inference for small models
  • Image generation
  • Development and testing

Request

Request a quote

Tell us the GPU, how many, and the term you need.