Europe
Availability on request
Price on requestConfiguration and region to be confirmed
GLOBALATLAS AI · L40S
NVIDIA L40S · 48 GB
48 GB provides room for larger working sets than 24 GB cards. It does not establish token throughput or the number of concurrent users.
Ada Lovelace48GB VRAM
- Memory technology
- GDDR6 ECC
- Workloads
- Generative AI, rendering and video workloads
Generative AI alongside visual workloads
- 01
48 GB for the complete working set
- 02
AI runtime or rendering engine compatibility
- 03
Separate measurements for generation and rendering
L40S offers a larger memory class than 24 GB accelerators. Keep L40 and L40S distinct when requesting a configuration.
Regional configurations
USA
Availability on request
Price on requestConfiguration and region to be confirmed
Asia
Availability on request
Price on requestConfiguration and region to be confirmed
Choosing L40S
Can I use an L40 measurement for L40S?
Treat them as distinct models. Repeat the test with the same software, scene or model, precision and batch settings on the selected hardware.
Technical references
Specification 1 →YOUR CONFIGURATION
Request an infrastructure quote
Choose resources and operating time. We will confirm a price for your configuration.
Each resource has its own listed configuration. Multiple GPU servers do not form an NVLink cluster. Monthly rates cover the full billing period; storage and traffic outside the configuration are not included.