Colfax Experience Center is offering a test drive program that provides you remote access to a virtual private server with an NVIDIA L40S GPU for evaluation. It is free of charge and available worldwide.
About the NVIDIA L40S GPU
Powered by the NVIDIA Ada Lovelace architecture, the universal NVIDIA L40S GPUs deliver unparalleled acceleration for generative AI, model training and inference as well as for 3D graphics, rendering, and video applications.
Unprecedented universal performance for the next generation of AI-enabled applications is made possible by a combination of:
- NVIDIA Fourth-Generation Tensor cores delivering up to 1,466 TFLOPS for deep learning tasks;
- NVIDIA Third-Generation RT Cores capable of up to 212 TFLOPS for ray tracing; and
- Single-precision NVIDIA CUDA® cores with up to 91.6 TFLOPS performance.
NVIDIA L40S GPUs make AI training 1.7x faster and generative AI performance 1.2x higher than with NVIDIA A100 GPUs. The transformer engine intelligently scans the layers of transformer architecture neural networks and automatically recasts between FP8 and FP16 precision to deliver faster AI performance and accelerate training and inference.
The L40S GPU is optimized for 24/7 enterprise data center operations and designed, built, tested and supported by NVIDIA to ensure maximum performance, durability, and uptime. The L40S GPU meets the latest data center standards, is NEBS Level 3 ready, and features secure boot with root trust technology, providing an additional layer of security for data centers.
- Datasheet (external link)
- Product Brief (external link)
- NVIDIA L40S on Cloud and Data Center page (external link)
Click to View Technical Specifications
| GPU Architecture | NVIDIA Ada Lovelace Architecture |
| GPU Memory | 48 GB GDDR6 with ECC |
| Memory Bandwidth | 864 GB/s |
| Interconnect Interface | PCIe Gen4 x16: 64 GB/s bidirectional |
| NVIDIA Ada Lovelace Architecture-Based CUDA® Cores | 18,176 |
| NVIDIA Third-Generation RT Cores | 142 |
| NVIDIA Fourth-Generation Tensor Cores | 568 |
| RT Core Performance TFLOPS | 568 |
| FP32 TFLOPS | 91.6 |
| TF32 Tensor Core TFLOPS | 183 | 366* |
| BFLOAT16 Tensor Core TFLOPS | 362.05 | 733 |
| FP16 Tensor Core | 362.05 | 733 |
| FP8 Tensor Core | 733 | 1,466* |
| Peak INT8 Tenor TOPS | 733 | 1,466* |
| Peak INT4 Tensor TOPS | 733 | 1,466* |
* With sparsity
How the Test Drive Works
Step 1
Complete the registration form
Step 2
If you qualify, you will receive an email message with connection instructions.
Step 3
Run your GPU-accelerated HPC or AI application
The Colfax GPU Test Drive program is intended for evaluation purposes only. Commercial use of computing time on any Colfax Test Drive system is strictly prohibited. We reserve the right to remove any user’s remote access at all times.