Ada Lovelace GPU
NVIDIA L4
24GB GDDR6 | 242 TFLOPS FP16 | 300 GB/s bandwidth
From $0.350 /hr
Inference Video transcoding Light AI workloads
Specifications
VRAM
24 GB GDDR6
FP16
242 TFLOPS
FP8
485 TFLOPS
FP4
N/A
Memory BW
300 GB/s
TDP
72W
Interconnect
PCIe Gen4
Architecture
Ada Lovelace
Cloud GPU Pricing (4 offers)
| Provider | Instance Type | vCPUs | RAM | Price/hr | Price/mo | Spot Price | Availability | Action |
|---|---|---|---|---|---|---|---|---|
Vast.aiCheapest | l4_24gb | 8 | 32 GB | $0.350/hr | $255.50/mo | $0.260-25.7% | Available | Rent on Vast.ai |
RunPod | l4-24gb | 8 | 62 GB | $0.390/hr | $284.70/mo | $0.270-30.8% | Available | Deploy on RunPod |
Google Cloud Platform | g2-standard-8 | 8 | 32 GB | $0.520/hr | $379.60/mo | $0.210-59.6% | Available | Deploy on Google Cloud |
Google Cloud Platform | g2-standard-8 | 8 | 32 GB | $0.570/hr | $416.10/mo | $0.230-59.7% | Available | Deploy on Google Cloud |
Last updated 57h ago
Compare With Other GPUs
Frequently Asked Questions
What is the cheapest NVIDIA L4 cloud provider?
The cheapest NVIDIA L4 is available on Vast.ai at $0.350/hr (l4_24gb).
How much does NVIDIA L4 cost per hour?
NVIDIA L4 cloud GPU pricing ranges from $0.350/hr to $0.570/hr depending on the provider and configuration.
What are the specs of NVIDIA L4?
NVIDIA L4 features 24GB GDDR6 memory, 242 TFLOPS FP16 performance, 300 GB/s memory bandwidth, and 72W TDP. Architecture: Ada Lovelace.
Is NVIDIA L4 good for AI training?
NVIDIA L4 is primarily designed for: Inference, Video transcoding, Light AI workloads. For large-scale training, consider higher-tier GPUs.