NVIDIA · Ada Lovelace Architecture

Rent NVIDIA L4 in the Cloud

Low-power inference accelerator. Excellent performance-per-watt for deploying AI models in production.

VRAM 24 GB GDDR6
Bandwidth 300 GB/s
FP16 121.0 TFLOPS
FP32 30.3 TFLOPS
TDP 72W
Architecture Ada Lovelace
Cheapest On-Demand $0.39/hr
Average On-Demand $0.39/hr
Providers 1

Compare NVIDIA L4 Cloud Pricing — 1 Providers

On-Demand

NVIDIA L4 — On-Demand pricing
Provider Price / GPU / hr Availability Notes Action
RunPod United States $0.39/hr CHEAPEST Available Secure Cloud Visit Provider

Prices last verified: April 13, 2026

NVIDIA L4 Technical Specifications

Manufacturer NVIDIA
Architecture Ada Lovelace
VRAM 24 GB GDDR6
Memory Bandwidth 300 GB/s
FP16 (Tensor) 121.0 TFLOPS
FP32 30.3 TFLOPS
TDP 72W
Release Year 2023
Segment Data center
Memory Type GDDR6

Best For

Inference video transcoding lightweight AI workloads

Compare with Other GPUs

See how NVIDIA L4 stacks up against other popular cloud GPUs in specs, pricing, and availability.

NVIDIA L4 vs NVIDIA H100 SXM NVIDIA L4 vs NVIDIA H200 SXM NVIDIA L4 vs NVIDIA A100 SXM (80GB) NVIDIA L4 vs NVIDIA L40S