NVIDIA L4 GPU cloud prices
Efficient, low-power Ada card for inference and video.
Compare vs other GPUs →Not enough price history yet — the trend fills in as we log more.
Provider-balanced managed-cloud median per GPU-hour (USD) · live data
Key Specifications
ArchitectureAda Lovelace
Memory24GB GDDR6
Memory Bandwidth300 GB/s
ReleaseQ1 2023
Compare Cloud Provider Prices
We’re tracking 19 providers for the NVIDIA L4 — on-demand runs $0.26 to $5.77/hr per GPU, with interruptible spot from $0.20/hr.
| Provider | Instance | Location | VRAM | Billing | $/GPU/hr | Availability | |
|---|---|---|---|---|---|---|---|
 AWSPublic pricing feed | 8× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.177Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.182Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.197Modeled | Not reported | View ↗ |
 Vast.aiMarketplace | 2× NVIDIA L4 | Czechia, CZ | 24GB | BURST · SPOT | $0.201 | Available | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | ap-northeast-2 | 24GB | BURST · SPOT | $0.213Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.219Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.222Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.224Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.225Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 4× NVIDIA L4 | ap-southeast-2 | 24GB | BURST · SPOT | $0.231Modeled | Not reported | View ↗ |
GOGPU OutletPublic pricing feed | 1× NVIDIA L4 | IN | 24GB | STANDARD | $0.26 | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.262Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.267Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | ap-south-1 | 24GB | BURST · SPOT | $0.279Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 4× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.282Modeled | Not reported | View ↗ |
 Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-northeast3-b | 24GB | BURST · SPOT | $0.295Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-northeast3-a | 24GB | BURST · SPOT | $0.295Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | ap-northeast-2 | 24GB | BURST · SPOT | $0.301Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-west-2 | 24GB | BURST · SPOT | $0.317Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 8× NVIDIA L4 | eu-west-2 | 24GB | BURST · SPOT | $0.322Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 4× NVIDIA L4 | eu-north-1 | 24GB | BURST · SPOT | $0.336Modeled | Not reported | View ↗ |
GOGPU OutletPublic pricing feed | 1× NVIDIA L4 | IS | 24GB | STANDARD | $0.34 | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | us-east4-c | 24GB | BURST · SPOT | $0.342Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | us-east4-a | 24GB | BURST · SPOT | $0.342Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-west-3 | 24GB | BURST · SPOT | $0.342Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-northeast3-a | 24GB | BURST · SPOT | $0.356Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-northeast3-b | 24GB | BURST · SPOT | $0.356Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-south1-c | 24GB | BURST · SPOT | $0.357Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-south1-b | 24GB | BURST · SPOT | $0.357Modeled | Not reported | View ↗ |
| Google CloudPublic pricing feed | 1× NVIDIA L4 | asia-south1-a | 24GB | BURST · SPOT | $0.357Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-west-3 | 24GB | BURST · SPOT | $0.358Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.359Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-west-3 | 24GB | BURST · SPOT | $0.364Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | ap-southeast-2 | 24GB | BURST · SPOT | $0.378Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 8× NVIDIA L4 | sa-east-1 | 24GB | BURST · SPOT | $0.385Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | eu-west-2 | 24GB | BURST · SPOT | $0.387Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 4× NVIDIA L4 | ap-south-1 | 24GB | BURST · SPOT | $0.392Modeled | Not reported | View ↗ |
| AWSPublic pricing feed | 1× NVIDIA L4 | ap-south-1 | 24GB | BURST · SPOT | $0.393Modeled | Not reported | View ↗ |
GOGPU OutletPublic pricing feed | 1× NVIDIA L4 | CZ | 24GB | STANDARD | $0.4 | Not reported | View ↗ |
| Vast.aiMarketplace | 2× NVIDIA L4 | Czechia, CZ | 24GB | STANDARD | $0.401 | Available | View ↗ |
Frequently Asked Questions
Why choose the NVIDIA L4?
24GB and a 72W envelope make it a cheap, dense choice for inference and transcoding.
When is the NVIDIA L4 not a good fit?
Too little compute/memory for training beyond small models.
How much does the NVIDIA L4 cost per hour?
NVIDIA L4 on-demand pricing currently runs $0.26/hr to $5.77/hr per GPU, depending on the provider and instance type. Interruptible spot capacity starts around $0.20/hr.
How much does the NVIDIA L4 cost per month?
At 720 hours per month, one NVIDIA L4 runs about $187 to $4,157 per month on demand. Lower spot rates are currently listed.
Which cloud providers offer the NVIDIA L4?
We currently track 1172 NVIDIA L4 listings across 19 providers. Pricing and availability vary by region and billing model.
Alternatives to NVIDIA L4
NVIDIA A100 40GB
Previous-generation workhorse — more affordable, widely available.
From $0.27 /hr · 27 providers
NVIDIA RTX 4090
Ada consumer flagship — strong price/performance for single-GPU work.
From $0.25 /hr · 24 providers
NVIDIA RTX A6000
Ampere workstation card with 48GB for inference and rendering.
From $0.27 /hr · 23 providers