Amazon Web Services is the world's largest cloud provider with 200+ services across compute, storage, databases, ML, and networking. Dominates in enterprise with the broadest global region footprint and the deepest service catalog, but pricing complexity and egress fees add up at scale.
NVIDIA L4 cloud pricing
- Vendor
- NVIDIA
- VRAM
- 24 GB
- Architecture
- Ada Lovelace
- FP16
- 121 TFLOPS
- Launched
- 2023
Lowest
$0.280
Median
$0.523
Highest
$0.805
5 results
| Provider | Plan | Price | Regions | Visit | ||||
|---|---|---|---|---|---|---|---|---|
|
|
G2 (1x L4) On-demand | 1 | 24 GB | 4 | 16 GB |
$0.700
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
G2 (1x L4) Spot | 1 | 24 GB | 4 | 16 GB |
$0.280
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
G2 (1x L4) Reserved | 1 | 24 GB | 4 | 16 GB |
$0.455
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
G6 (1x L4) On-demand | 1 | 24 GB | 4 | 16 GB |
$0.805
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
G6 (1x L4) Reserved | 1 | 24 GB | 4 | 16 GB |
$0.523
/GPU-hr
Verified
|
3 countries | Visit → |
Providers offering this GPU
Google Cloud Platform combines world-class data analytics, AI infrastructure (TPUs, Vertex AI), and the original managed Kubernetes. Its global fiber backbone and Preemptible VMs offer compelling price-performance for data-heavy and containerized workloads.
Baseten is a production model-serving platform with built-in autoscaling, per-minute GPU billing, and SOC 2/HIPAA compliance. Designed for teams deploying LLMs and diffusion models at scale, it handles cold starts, traffic spikes, and infrastructure tuning so engineers can focus on model quality rather than platform reliability.
Frequently asked questions
How much does an NVIDIA L4 cost per hour in the cloud?
The lowest on-demand NVIDIA L4 price we track is $0.280 per GPU-hour. Spot and reserved rates are usually lower; sort the table above by price to see the current rate from every provider.
What is the cheapest NVIDIA L4 cloud provider?
Sort the table by price (low to high) to see the cheapest NVIDIA L4 provider right now. Marketplace and spot providers often undercut hyperscalers by a wide margin for the same NVIDIA L4.
Which cloud providers offer NVIDIA L4 GPUs?
Every provider with published NVIDIA L4 availability is listed above, with per-hour pricing, the number of GPUs per instance, region coverage, and on-demand, spot, and reserved rates.
Is spot NVIDIA L4 cheaper than on-demand?
Yes. Spot (preemptible) capacity is typically 40-70% cheaper than on-demand but can be reclaimed at short notice. Use the pricing-mode filter to compare on-demand, spot, and reserved rows side by side.
How much VRAM does the NVIDIA L4 have?
The NVIDIA L4 ships with 24 GB of VRAM. Larger VRAM lets you fit bigger models and batch sizes without sharding.
Is the NVIDIA L4 good for AI training and inference?
The NVIDIA L4 is used for both LLM training and inference. Match its VRAM and throughput (shown above) to your model size, and use spot capacity for fault-tolerant training to cut costs.