NVIDIAProfessional GPU
RTX 6000 Ada
Workstation · Inference · HPC
- Memory
- 48 GB GDDR6 ECC
- Bandwidth
- 960 GB/s
- Form factor
- PCIe Gen4 dual-slot
Ada Lovelace PCIe accelerator with 48 GB GDDR6 ECC, for inference, rendering and virtualisation.
The L40S is a dual-slot PCIe data-center GPU aimed at inference, graphics and mixed workloads rather than large-scale training. Because it is a standard PCIe card with passive cooling, it fits conventional 2U and 4U servers without an HGX baseboard, which makes it a common choice for organisations adding accelerated capacity to existing infrastructure.
InferenceL40S is sourced for inference deployments within gpus, configured to the host platform and quantity specified in your requirement.
VirtualisationL40S is sourced for virtualisation deployments within gpus, configured to the host platform and quantity specified in your requirement.
WorkstationL40S is sourced for workstation deployments within gpus, configured to the host platform and quantity specified in your requirement.
Passively cooled — confirm chassis airflow and power headroom for the intended number of cards per server.
Workstation · Inference · HPC
Tell us your requirement and we'll reply with sourcing options and lead time.