L40S GPU
The Nvidia L40S provides multi-workload acceleration for large language model (LLM) inference and training, graphics, and video applications. Based on the latest Ada Lovelace architecture.
Starting Price
$0.67/hr
Available on 7 cloud providers
Key Specifications
💾Memory
48GB VRAM
🏗️Architecture
Ada Lovelace
⚙️Compute Units
N/A
🧮Tensor Cores
568
Technical Specifications
Hardware Details
ManufacturerNVIDIA
ArchitectureAda Lovelace
CUDA Cores18176
Tensor Cores568
RT Cores142
Compute UnitsN/A
GenerationN/A
Memory & Performance
VRAM48GB
Memory Interface384-bit
Memory Bandwidth864 GB/s
FP32 Performance91.6 TFLOPS
FP16 Performance362.05 TFLOPS
INT8 Performance733 TOPS
Performance
Computing Power
CUDA Cores18,176
Tensor Cores568
RT Cores142
Computational Performance
FP32 (TFLOPS)91.6
FP16 (TFLOPS)362.05
INT8 (TOPS)733
Common Use Cases
generative AI, model training and inference, rendering, metaverse apps
Machine Learning & AI
- Training large language models and transformers
- Computer vision and image processing
- Deep learning model development
- High-performance inference workloads
Graphics & Compute
- 3D rendering and visualization
- Scientific simulations
- Data center graphics virtualization
- High-performance computing (HPC)
Market Context
The L40S sits within NVIDIA's Ada Lovelace architecture lineup, positioned in the high performance tier.
Cloud Availability
Available across 7 cloud providers with prices ranging from $0.67/hr. Pricing and availability may vary by region and provider.
Market Position
Released in 2023, this GPU is positioned for professional workloads.