NVIDIA L40S
0
CUDA Cores
48GB
VRAM
864
GB/s
Technical Specifications
0
CUDA Cores
0
Base MHz
0
Boost MHz
48GB GDDR6
384-bit bus
Performance
90
FP32 TFLOPS
180
FP16 TFLOPS
350W
TDP
Cloud Availability
6
Available Instances
$0.96/hr
Starting Price
Detailed Specifications
| Architecture | Ada Lovelace (Unknown) |
| Release Date | 2023-10-01 |
| Launch Price | $2,000.00 |
| Process | 4nm |
| Transistors |
AI Features
Gen 4
Tensor Cores
Enabled
Transformer Engine
Not Supported
Flash Attention
Physical Specifications
Dimensions
10.5in
Length
4.4in
Width
2-slot
Height
About L40S GPU
The NVIDIA L40S was introduced in 2023 on the Ada Lovelace architecture. Its specification combines its vendor-specific compute units with 48GB of GDDR6, which defines both the workloads it can run and the batch sizes it can hold.
Based on its 48GB memory capacity, the L40S is best evaluated for model fine-tuning, production inference, image generation, and multi-GPU development. Always size the model, optimizer state, KV cache, and framework overhead together rather than choosing a GPU from core count alone.
GPUvec currently tracks 6 cloud listings for the L40S, with the lowest observed hourly rate at $0.96. Prices and availability can change by provider, region, and instance configuration, so verify the final rate before starting a workload.
For software planning, the L40S combines 864 GB/s of memory bandwidth, CUDA compute capability 8.9, and a 4nm manufacturing process. Confirm the minimum CUDA, PyTorch, TensorFlow, or driver version required by your application before renting or purchasing hardware.
Rent L40S from Our Partners
Get started quickly with these trusted GPU cloud providers. We may earn a commission when you sign up.
External Resources
Learn more about GPUs from these authoritative sources:
Official CUDA programming guide
NVIDIA GPU Specifications →Official NVIDIA GPU specs
TechPowerUp GPU Database →Comprehensive GPU specifications
CUDA Compute Capability Guide →GPU compute capability reference
What You Need to Know About the L40S
Complete Specifications for the NVIDIA L40S
The NVIDIA L40S uses the Ada Lovelace architecture and combines vendor-specific compute units with 48GB of GDDR6. GPUvec records 864 GB/s memory bandwidth alongside architecture, release date, process node, accelerator features, and physical specifications so the page can be evaluated from source data instead of model-name assumptions.
Compare NVIDIA L40S Cloud Rental Prices per Hour
GPUvec currently tracks 6 cloud listings, with an observed starting rate of $0.96 per hour. Provider rates can vary by region, host configuration, billing model, and availability. Use the provider directory and related-GPU links to compare an available alternative when this exact accelerator is not listed.
Is the NVIDIA L40S the Right GPU for Your AI Workload?
Start with the 48GB memory limit, then evaluate 864 GB/s memory bandwidth, compute features, and software compatibility for your workload. Model weights, KV cache, activations, optimizer state, and framework overhead all consume memory. Compare this page with related accelerators rather than assuming that a higher core count alone guarantees a better training or inference result.
Top GPUs for Training and Inference
| Category | Rank 1 | Rank 2 | Rank 3 |
|---|---|---|---|
| Best for Training | NVIDIA H200 | NVIDIA H100 | NVIDIA B200 |
| Best for Inference | NVIDIA A40 | NVIDIA A100 | NVIDIA A10 |
Compare GPU specifications and cloud instances to find the best GPU for your workload.